diff --git a/latest/.buildinfo b/latest/.buildinfo index 40066c9e5f..e399b071ba 100644 --- a/latest/.buildinfo +++ b/latest/.buildinfo @@ -1,4 +1,4 @@ # Sphinx build info version 1 # This file hashes the configuration used when building these files. When it is not found, a full rebuild will be done. -config: 12c1352bd1428d2c6ac709024163b9d8 +config: 5c850ce0a6f2d0ce79a91d25fbeeb241 tags: 645f666f9bcd5a90fca523b33c5a78b7 diff --git a/latest/_cpp_gen/executor.html b/latest/_cpp_gen/executor.html index 20c88f06f3..f1700a377d 100644 --- a/latest/_cpp_gen/executor.html +++ b/latest/_cpp_gen/executor.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -357,6 +358,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -421,6 +423,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -455,6 +458,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1250,6 +1254,553 @@ + +
    +

    transferAgent.h#

    +
    +
    +namespace tensorrt_llm
    +
    +
    +namespace executor
    +
    +
    +namespace kv_cache#
    +
    +

    Typedefs

    +
    +
    +using TransferDescs = MemoryDescs#
    +
    + +
    +
    +using RegisterDescs = MemoryDescs#
    +
    + +
    +
    +using SyncMessage = std::string#
    +
    + +
    +
    +using ConnectionInfoType = std::string#
    +
    + +
    +
    +

    Enums

    +
    +
    +enum class MemoryType : uint8_t#
    +

    Values:

    +
    +
    +enumerator kDRAM#
    +
    + +
    +
    +enumerator kVRAM#
    +
    + +
    +
    +enumerator kBLK#
    +
    + +
    +
    +enumerator kOBJ#
    +
    + +
    +
    +enumerator kFILE#
    +
    + +
    + +
    +
    +enum class TransferOp : uint8_t#
    +

    Values:

    +
    +
    +enumerator kREAD#
    +
    + +
    +
    +enumerator kWRITE#
    +
    + +
    + +
    +
    +

    Functions

    +
    +
    +template<typename ...Args>
    std::unique_ptr<BaseTransferAgent> makeTransferAgent( + +
    +
    std::string const &backend,
    +
    Args&&... args,
    +
    + +)#
    +
    + +
    +
    +
    +class AgentDesc#
    +
    +

    Public Functions

    +
    +
    +inline AgentDesc(std::string backendAgentDesc)#
    +
    + +
    +
    +inline std::string const &getBackendAgentDesc() const noexcept#
    +
    + +
    +
    +

    Private Members

    +
    +
    +std::string mBackendAgentDesc#
    +
    + +
    +
    + +
    +
    +struct BaseAgentConfig#
    +
    +

    Public Members

    +
    +
    +std::string mName#
    +
    + +
    +
    +bool useProgThread#
    +
    + +
    +
    + +
    +
    +class BaseTransferAgent#
    +
    +

    Public Functions

    +
    +
    +virtual ~BaseTransferAgent() = default#
    +
    + +
    +
    +virtual void registerMemory(RegisterDescs const &descs) = 0#
    +
    + +
    +
    +virtual void deregisterMemory(RegisterDescs const &descs) = 0#
    +
    + +
    +
    +virtual void loadRemoteAgent( + +
    +
    std::string const &name,
    +
    AgentDesc const &agentDesc,
    +
    + +) = 0#
    +
    + +
    +
    +virtual AgentDesc getLocalAgentDesc() = 0#
    +
    + +
    +
    +virtual void invalidateRemoteAgent(std::string const &name) = 0#
    +
    + +
    +
    +virtual std::unique_ptr<TransferStatus> submitTransferRequests( + +
    +
    TransferRequest const &request,
    +
    + +) = 0#
    +
    + +
    +
    +virtual void notifySyncMessage( + +
    +
    std::string const &name,
    +
    SyncMessage const &syncMessage,
    +
    + +) = 0#
    +
    + +
    +
    +virtual std::unordered_map<std::string, std::vector<SyncMessage>> getNotifiedSyncMessages( + +
    +
    + +) = 0#
    +
    + +
    +
    +virtual ConnectionInfoType getConnectionInfo() = 0#
    +
    + +
    +
    +virtual void connectRemoteAgent( + +
    +
    std::string const &name,
    +
    ConnectionInfoType const &connectionInfo,
    +
    + +) = 0#
    +
    + +
    +
    +virtual bool checkRemoteDescs( + +
    +
    std::string const &name,
    +
    MemoryDescs const &memoryDescs,
    +
    + +) = 0#
    +
    + +
    +
    + +
    +
    +class DynLibLoader#
    +
    +

    Public Functions

    +
    +
    +void *getHandle(std::string const &name)#
    +
    + +
    +
    +template<typename FunctionT>
    inline FunctionT getFunctionPointer( + +
    +
    std::string const &libName,
    +
    std::string const &funcName,
    +
    + +)#
    +
    + +
    +
    +~DynLibLoader()#
    +
    + +
    +
    +DynLibLoader() = default#
    +
    + +
    +
    +DynLibLoader(DynLibLoader const&) = delete#
    +
    + +
    +
    +DynLibLoader &operator=(DynLibLoader const&) = delete#
    +
    + +
    +
    +

    Public Static Functions

    +
    +
    +static DynLibLoader &getInstance()#
    +
    + +
    +
    +

    Private Members

    +
    +
    +std::mutex mDllMutex#
    +
    + +
    +
    +std::unordered_map<std::string, void*> mHandlers#
    +
    + +
    +
    +

    Private Static Functions

    +
    +
    +static void *dlSym(void *handle, char const *symbol)#
    +
    + +
    +
    + +
    +
    +class MemoryDesc#
    +
    +

    Public Functions

    +
    +
    +inline MemoryDesc( + +
    +
    std::vector<char> const &vec,
    +
    uint32_t deviceId = 0,
    +
    + +)#
    +
    + +
    +
    +inline MemoryDesc(void *addr, size_t len, uint32_t deviceId)#
    +
    + +
    +
    +inline MemoryDesc(uintptr_t addr, size_t len, uint32_t deviceId)#
    +
    + +
    +
    +inline uintptr_t getAddr() const noexcept#
    +
    + +
    +
    +inline size_t getLen() const noexcept#
    +
    + +
    +
    +inline uint32_t getDeviceId() const noexcept#
    +
    + +
    +
    +

    Public Static Functions

    +
    +
    +static void serialize(MemoryDesc const &memoryDesc, std::ostream &os)#
    +
    + +
    +
    +static MemoryDesc deserialize(std::istream &is)#
    +
    + +
    +
    +static size_t serializedSize(MemoryDesc const &memoryDesc)#
    +
    + +
    +
    +

    Private Members

    +
    +
    +uintptr_t mAddr#
    +
    + +
    +
    +size_t mLen#
    +
    + +
    +
    +uint32_t mDeviceId#
    +
    + +
    +
    + +
    +
    +class MemoryDescs#
    +
    +

    Public Functions

    +
    +
    +inline MemoryDescs(MemoryType type, std::vector<MemoryDesc> descs)#
    +
    + +
    +
    +inline MemoryType getType() const noexcept#
    +
    + +
    +
    +inline std::vector<MemoryDesc> const &getDescs() const noexcept#
    +
    + +
    +
    +

    Private Members

    +
    +
    +MemoryType mType#
    +
    + +
    +
    +std::vector<MemoryDesc> mDescs#
    +
    + +
    +
    + +
    +
    +class TransferRequest#
    +
    +

    Public Functions

    +
    +
    +inline TransferRequest( + +
    +
    TransferOp op,
    +
    TransferDescs srcDescs,
    +
    TransferDescs dstDescs,
    +
    std::string const &remoteName,
    +
    std::optional<SyncMessage> syncMessage = std::nullopt,
    +
    + +)#
    +
    + +
    +
    +inline TransferOp getOp() const noexcept#
    +
    + +
    +
    +inline TransferDescs const &getSrcDescs() const noexcept#
    +
    + +
    +
    +inline TransferDescs const &getDstDescs() const noexcept#
    +
    + +
    +
    +inline std::string const &getRemoteName() const noexcept#
    +
    + +
    +
    +inline std::optional<SyncMessage> getSyncMessage() const noexcept#
    +
    + +
    +
    +

    Private Members

    +
    +
    +TransferOp mOp#
    +
    + +
    +
    +TransferDescs mSrcDescs#
    +
    + +
    +
    +TransferDescs mDstDescs#
    +
    + +
    +
    +std::string mRemoteName#
    +
    + +
    +
    +std::optional<SyncMessage> mSyncMessage#
    +
    + +
    +
    + +
    +
    +class TransferStatus#
    +
    +

    Public Functions

    +
    +
    +virtual ~TransferStatus() = default#
    +
    + +
    +
    +virtual bool isCompleted() const = 0#
    +
    + +
    +
    +virtual void wait() const = 0#
    +
    + +
    +
    + +
    + +
    + +
    +

    serialization.h#

    @@ -1514,6 +2065,28 @@ static size_t serializedSize(kv_cache::SocketState const &state)#
    +
    +
    +static kv_cache::AgentState deserializeAgentState(std::istream &is)#
    +
    + +
    +
    +static void serialize( + +
    +
    kv_cache::AgentState const &state,
    +
    std::ostream &os,
    +
    + +)#
    +
    + +
    +
    +static size_t serializedSize(kv_cache::AgentState const &state)#
    +
    +
    static kv_cache::CacheState deserializeCacheState(std::istream &is)#
    @@ -2398,6 +2971,34 @@ )#
    +
    +
    +static SpecDecodingStats deserializeSpecDecodingStats( + +
    +
    std::istream &is,
    +
    + +)#
    +
    + +
    +
    +static void serialize( + +
    +
    SpecDecodingStats const &specDecStats,
    +
    std::ostream &os,
    +
    + +)#
    +
    + +
    +
    +static size_t serializedSize(SpecDecodingStats const &specDecStats)#
    +
    +
    static IterationStats deserializeIterationStats( @@ -2620,8 +3221,8 @@
    -
    -namespace kv_cache#
    +
    +namespace kv_cache
    @@ -2649,6 +3250,11 @@ using SizeType32 = std::int32_t#
    +
    +
    +using SizeType64 = std::int64_t#
    +
    +
    using FloatType = float#
    @@ -3045,6 +3651,31 @@
    +
    +
    +enum class KvCacheTransferMode#
    +

    Enum describing the transfer mode for KV cache.

    +

    Values:

    +
    +
    +enumerator DRAM#
    +

    Copy to/from CPU memory (original approach).

    +
    + +
    +
    +enumerator GDS#
    +

    Attempt GPUDirect Storage (cuFile).

    +
    + +
    +
    +enumerator POSIX_DEBUG_FALLBACK#
    +

    Force a POSIX read/write for debugging.

    +
    + +
    +

    Functions

    @@ -3787,6 +4418,12 @@

    Stats specific to inflight batching.

    +
    +
    +std::optional<SpecDecodingStats> specDecStats#
    +

    Stats specific to speculative decoding.

    +
    +
    @@ -4133,6 +4770,53 @@ +
    +
    +struct SpecDecodingStats#
    +
    +#include <types.h>
    +

    Struct that holds speculative decoding stats.

    +
    +

    Public Members

    +
    +
    +SizeType64 numDraftTokens#
    +

    Total number of proposed draft tokens for all requests.

    +
    + +
    +
    +SizeType64 numAcceptedTokens#
    +

    Total number of accepted draft tokens for all requests.

    +
    + +
    +
    +SizeType64 numRequestsWithDraftTokens#
    +

    Number of requests with at least one draft token in batch.

    +
    + +
    +
    +double acceptanceLength#
    +

    Acceptance length, defined as average number of tokens produced per step for all requests with at least one draft token.

    +
    + +
    +
    +double iterLatencyMS#
    +

    Iteration latency for draft token generation only (ms)

    +
    + +
    +
    +double draftOverhead#
    +

    Draft overhead, defined as iterLatencyMS (specdec) / iterLatencyMS (total)

    +
    + +
    +
    +
    struct StaticBatchingStats#
    @@ -6909,16 +7593,18 @@
    -
    -explicit KvCacheRetentionConfig( +
    +explicit KvCacheRetentionConfig(
    std::vector<TokenRangeRetentionConfig> const &tokenRangeRetentionPriorities,
    RetentionPriority decodeRetentionPriority = kDefaultRetentionPriority,
    std::optional<std::chrono::milliseconds> decodeDurationMs = std::nullopt,
    +
    KvCacheTransferMode transferMode = KvCacheTransferMode::DRAM,
    +
    std::optional<std::string> directory = std::nullopt,
    -)#
    +)#
    @@ -6946,6 +7632,16 @@ ) const#
    +
    +
    +KvCacheTransferMode getTransferMode() const#
    +
    + +
    +
    +std::optional<std::string> getDirectory() const#
    +
    +
    std::vector<RetentionPriorityAndDuration> getPerBlockRetentionPriorityDuration( @@ -7003,6 +7699,18 @@

    The duration in ms that decode blocks should remain at their assigned priority level.

    +
    +
    +KvCacheTransferMode mTransferMode#
    +

    The transfer mode for the block.

    +
    + +
    +
    +std::optional<std::string> mDirectory#
    +

    Name of the directory if transfer mode is GDS or POSIX_DEBUG_FALLBACK.

    +
    +
    @@ -9796,7 +10504,48 @@
    namespace kv_cache
    -
    +
    +
    +struct AgentState#
    +
    +

    Public Functions

    +
    +
    +inline AgentState(std::string agentName, std::string connectionInfo)#
    +
    + +
    +
    +AgentState() = default#
    +
    + +
    +
    +inline bool operator==(AgentState const &other) const noexcept#
    +
    + +
    +
    +inline std::string toString() const#
    +
    + +
    +
    +

    Public Members

    +
    +
    +std::string mAgentName#
    +
    + +
    +
    +std::string mConnectionInfo#
    +
    + +
    +
    + +
    class CacheState#
    @@ -10085,6 +10834,18 @@ inline CommState(std::uint16_t port, std::string ip)#
    +
    +
    +inline explicit CommState( + +
    +
    std::vector<AgentState> agentState,
    +
    int selfIdx = -1,
    +
    + +)#
    +
    +
    inline bool isMpiState() const noexcept#
    @@ -10095,6 +10856,11 @@ inline bool isSocketState() const noexcept#
    +
    +
    +inline bool isAgentState() const noexcept#
    +
    +
    inline MpiState const &getMpiState() const#
    @@ -10105,6 +10871,11 @@ inline std::vector<SocketState> const &getSocketState() const#
    +
    +
    +inline std::vector<AgentState> const &getAgentState() const#
    +
    +
    inline int getSelfIdx() const noexcept#
    @@ -10125,7 +10896,7 @@

    Private Members

    -std::variant<std::monostate, MpiState, std::vector<SocketState>> mState#
    +std::variant<std::monostate, MpiState, std::vector<SocketState>, std::vector<AgentState>> mState#
    @@ -10215,11 +10986,6 @@ namespace tensorrt_llm
    -namespace batch_manager
    -
    - -
    -
    namespace executor
    @@ -10496,6 +11262,112 @@
  • tensorrt_llm::runtime
    • +
  • transferAgent.h +
  • serialization.h
  • types.h
    • +
  • KvCacheTransferMode +
  • operator<<()
  • operator<<()
  • tensorrt_llm::executor::DebugTensorsPerIteration
  • tensorrt_llm::executor::KvCacheStats
    • +
  • tensorrt_llm::executor::SpecDecodingStats +
  • tensorrt_llm::executor::StaticBatchingStats
    • +
    +
    +inline SizeType32 getFirstLocalLayer( + +
    +
    SizeType32 pipelineParallelism = 1,
    +
    SizeType32 pipelineParallelismRank = 0,
    +
    + +) const#
    +
    +
    inline SizeType32 countLowerRankLayers( @@ -2204,8 +2220,15 @@
    -
    -inline SizeType32 getNbLayers(SizeType32 pipelineParallelism = 1) const#
    +
    +inline SizeType32 getNbLayers( + +
    +
    SizeType32 pipelineParallelism = 1,
    +
    SizeType32 pipelineParallelismRank = 0,
    +
    + +) const#
    @@ -11199,6 +11222,19 @@ one more than decoding draft tokens for prediction from primary head

    +
    +
    +TensorPtr getSequenceLengths(SizeType32 batchIdx) const#
    +
    +
    Parameters:
    +

    batchIdx – index of the batch

    +
    +
    Returns:
    +

    [maxBeamWidth], sequence lengths for request batchIdx, on gpu

    +
    +
    +
    +
    TensorPtr getAllNewTokens() const#
    @@ -11270,6 +11306,11 @@ one more than decoding draft tokens for prediction from primary head

    +
    +
    +SizeType32 getMaxBatchSize() const#
    +
    +
    SizeType32 getMaxBeamWidth() const#
    @@ -11500,6 +11541,11 @@ one more than decoding draft tokens for prediction from primary head

    TensorPtr mAllReduceCommPtrs#
    +
    +
    +TensorPtr mFlagPtrs#
    +
    +
    std::vector<runtime::IpcMemory> mIpcMemoryHandles#
    @@ -12171,8 +12217,9 @@ one more than decoding draft tokens for prediction from primary head

  • getVocabSize()
  • getVocabSizePadded()
  • countLocalLayers()
    • +
  • getFirstLocalLayer()
  • countLowerRankLayers()
    • -
  • getNbLayers()
    • +
  • getNbLayers()
  • getNbAttentionLayers()
  • getNbRnnLayers()
  • getNbHeads()
    • @@ -13526,6 +13573,7 @@ one more than decoding draft tokens for prediction from primary head

  • getLogProbs()
  • getLogProbs()
  • getSequenceLengths()
    • +
  • getSequenceLengths()
  • getAllNewTokens()
  • getNextDraftTokens()
  • getPrevDraftTokensLengths()
    • @@ -13533,6 +13581,7 @@ one more than decoding draft tokens for prediction from primary head

  • getAcceptedLengthsCumSum()
  • getAcceptedPackedPaths()
  • getFinishedSteps()
    • +
  • getMaxBatchSize()
  • getMaxBeamWidth()
  • getMaxSequenceLength()
  • getMaxDecodingDecoderTokens()
    • @@ -13566,6 +13615,7 @@ one more than decoding draft tokens for prediction from primary head

  • TensorPtr
  • AllReduceBuffers()
  • mAllReduceCommPtrs
    • +
  • mFlagPtrs
  • mIpcMemoryHandles
    • @@ -13717,6 +13767,15 @@ one more than decoding draft tokens for prediction from primary head

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_downloads/b509390ba70e52fabb10dbd9d15d5118/attention.py b/latest/_downloads/b509390ba70e52fabb10dbd9d15d5118/attention.py new file mode 100644 index 0000000000..32dcea9fff --- /dev/null +++ b/latest/_downloads/b509390ba70e52fabb10dbd9d15d5118/attention.py @@ -0,0 +1,1081 @@ +import math +import weakref +from enum import IntEnum +from typing import Optional, Union, cast + +import torch +from torch import nn + +from tensorrt_llm.mapping import Mapping + +from ..attention_backend import (AttentionInputType, AttentionMetadata, + TrtllmAttention, TrtllmAttentionMetadata) +from ..attention_backend.interface import (PositionalEmbeddingParams, + PredefinedAttentionMask) +from ..attention_backend.utils import create_attention, get_attention_backend +from ..distributed import AllReduceParams +from ..model_config import ModelConfig +from ..peft.lora.layer import LoraLayer, LoraModuleType +from ..utils import Fp4QuantizedTensor, get_model_extra_attrs +from .linear import Linear, TensorParallelMode, WeightMode, WeightsLoadingConfig +from .multi_stream_utils import maybe_execute_in_parallel +from .rms_norm import RMSNorm +from .rotary_embedding import RotaryEmbedding + + +class QkNormType(IntEnum): + """ + The type of QK normalization. + """ + none = 0 # No normalization applied to Q and K + pre_rope = 1 # Apply normalization before Rope + post_rope = 2 # Apply normalization after Rope + + +class Attention(nn.Module): + + def __init__( + self, + *, + hidden_size: int, + num_attention_heads: int, + num_key_value_heads: int, + max_position_embeddings: int, + bias: bool, + pos_embd_params: Optional[PositionalEmbeddingParams] = None, + qk_norm_type: QkNormType = QkNormType.none, + layer_idx: Optional[int] = None, + dtype: torch.dtype = None, + dense_bias: Optional[bool] = None, + config: Optional[ModelConfig] = None, + q_scaling: float = 1.0, + attention_chunk_size: Optional[int] = None, + ): + """ + Initialize the Attention module. + + Args: + hidden_size (int): The size of the hidden dimension. + num_attention_heads (int): The number of attention heads. + num_key_value_heads (int): The number of key value heads. + max_position_embeddings (int): The maximum position embeddings. + bias (bool): Whether to use bias in the linear layers. + pos_embd_params (PositionalEmbeddingParams): The positional embedding parameters. + qk_norm_type (QkNormType): The type of QK normalization. + layer_idx (int): The layer index. + dtype (torch.dtype): The data type. + dense_bias (bool): Whether to use bias in the output projection layer. + config (ModelConfig): The model configuration. + q_scaling (float): The scaling factor for the qk_scale. The definition is $O = softmax(QK^T * qk_scale) * V, qk_scale = 1 / (sqrt(head_dim) * q_scaling)$. The default value is 1.0. + attention_chunk_size (int): See [Chunked Attention] below. + """ + super().__init__() + self.layer_idx = layer_idx + + config = config or ModelConfig() + self.hidden_size = hidden_size + self.num_heads = num_attention_heads + self.head_dim = getattr(config.pretrained_config, "head_dim", + self.hidden_size // self.num_heads) + self.num_key_value_heads = num_key_value_heads + self.num_key_value_groups = self.num_heads // self.num_key_value_heads + self.max_position_embeddings = max_position_embeddings + self.pos_embd_params = pos_embd_params + self.qk_norm_type = qk_norm_type + self.dense_bias = dense_bias + self.q_scaling = q_scaling + + # [Chunked Attention] + # Chunked attention is applied to context requests only. Chunked attention will be + # applied when this field is specified and mMaskType == CAUSAL. + # + # In chunked attention, we break context requests into chunks of a specified size. Tokens can only + # attend to tokens in the same chunk. So, for example, if the chunk size is 3, we might have a mask + # that looks like this: + # + # 1 0 0 0 0 0 + # 1 1 0 0 0 0 + # 1 1 1 0 0 0 + # 0 0 0 1 0 0 + # 0 0 0 1 1 0 + # 0 0 0 1 1 1 + self.attention_chunk_size = attention_chunk_size + + if dense_bias is None: + self.dense_bias = bias + + # tensor parallel + tp_size = config.mapping.tp_size + pp_size = config.mapping.pp_size + if config.mapping.enable_attention_dp: + tp_size = 1 + + mapping = Mapping( + world_size=tp_size * pp_size, + tp_size=tp_size, + pp_size=pp_size, + rank=config.mapping.rank, + gpus_per_node=config.mapping.gpus_per_node, + enable_attention_dp=config.mapping.enable_attention_dp, + ) + assert self.num_heads % tp_size == 0 + self.num_heads = self.num_heads // tp_size + self.num_key_value_heads = (self.num_key_value_heads + tp_size - + 1) // tp_size + self.q_size = self.num_heads * self.head_dim + self.kv_size = self.num_key_value_heads * self.head_dim + + self.qkv_proj = Linear( + self.hidden_size, + tp_size * self.q_size + 2 * tp_size * self.kv_size, + bias=bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.COLUMN, + weights_loading_config=WeightsLoadingConfig( + weight_mode=WeightMode.FUSED_QKV_LINEAR), + quant_config=config.get_quant_config(), + skip_create_weights_in_init=config.skip_create_weights_in_init, + ) + self.o_lora = LoraLayer([LoraModuleType.ATTENTION_DENSE], + [self.hidden_size]) + + self.o_proj = Linear( + tp_size * self.q_size, + self.hidden_size, + bias=self.dense_bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.ROW, + quant_config=config.get_quant_config(), + skip_create_weights_in_init=config.skip_create_weights_in_init, + lora=self.o_lora, + ) + + self.quant_config = config.get_quant_config() + self.attn_backend = config.attn_backend + attn_cls = get_attention_backend(self.attn_backend) + + # These two modules are mutually exclusive - either splitted_qkv_lora or fused_qkv_lora will be used, + # but never both at the same time. splitted_qkv_lora handles Q,K,V separately while fused_qkv_lora + # handles them as a single fused operation. + self.splitted_qkv_lora = LoraLayer([ + LoraModuleType.ATTENTION_Q, LoraModuleType.ATTENTION_K, + LoraModuleType.ATTENTION_V + ], [self.q_size, self.kv_size, self.kv_size]) + self.fused_qkv_lora = LoraLayer([LoraModuleType.ATTENTION_QKV], + [self.q_size + 2 * self.kv_size]) + + self.o_lora = LoraLayer([LoraModuleType.ATTENTION_DENSE], + [self.hidden_size]) + + # enable_rope_fusion: Whether to fuse RoPE into the attention OP. + # If true, RoPE will be applied in self.attn.forward. + # If false, RoPE will be applied in self.apply_rope. + self.enable_rope_fusion = attn_cls.support_fused_rope( + ) and self.qk_norm_type != QkNormType.post_rope + + self.rotary_emb = None + if not self.enable_rope_fusion and self.pos_embd_params is not None: + self.rotary_emb = RotaryEmbedding( + self.pos_embd_params.rope, + head_dim=self.head_dim, + is_neox=self.pos_embd_params.is_neox, + ) + + self.attn = create_attention( + self.attn_backend, + self.layer_idx, + self.num_heads, + self.head_dim, + self.num_key_value_heads, + pos_embd_params=self.pos_embd_params + if self.enable_rope_fusion else None, + quant_config=self.quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init, + q_scaling=self.q_scaling, + attention_chunk_size=self.attention_chunk_size, + ) + + self.support_fused_qkv = self.attn.support_fused_qkv() + + if not config.skip_create_weights_in_init: + self.create_weights() + + def create_weights(self): + # self.attn has no weights but has states that are related to quant_config, + # which could be modified after __init__ + self.attn.update_quant_config(self.quant_config) + + def split_qkv(self, q, k=None, v=None): + if k is None and v is None: + q, k, v = q.split([self.q_size, self.kv_size, self.kv_size], dim=-1) + return q, k, v + + def convert_qkv(self, q, k, v): + if k is None and v is None and not self.support_fused_qkv: + q, k, v = self.split_qkv(q) + elif k is not None and v is not None and self.support_fused_qkv: + qkv = torch.concat([q, k, v], dim=-1) + q, k, v = qkv, None, None + return q, k, v + + def forward( + self, + position_ids: Optional[torch.LongTensor], + hidden_states: Union[torch.Tensor, Fp4QuantizedTensor], + attn_metadata: AttentionMetadata, + attention_mask: PredefinedAttentionMask = PredefinedAttentionMask. + CAUSAL, + mrope_config: Optional[dict] = None, + all_reduce_params: Optional[AllReduceParams] = None, + lora_params: Optional[dict] = None, + attention_window_size: Optional[int] = None, + **kwargs, + ) -> torch.Tensor: + """ + Forward pass for the Attention module. + + Args: + position_ids (Optional[torch.LongTensor]): The position IDs. + hidden_states (torch.Tensor): The hidden states. + attn_metadata (AttentionMetadata): The attention metadata. + attention_mask (PredefinedAttentionMask): The attention mask type. + mrope_config (Optional[dict]): The MROPE configuration. + all_reduce_params (Optional[AllReduceParams]): The all reduce parameters. + lora_params (Optional[dict]): The LoRA parameters. + attention_window_size (Optional[int]): The attention window size. + + Returns: + torch.Tensor: The output tensor. + """ + qkv = self.qkv_proj(hidden_states) + + if bool(lora_params): + qkv_lora = self.splitted_qkv_lora(hidden_states, lora_params, + self.layer_idx) + if qkv_lora is not None: + qkv = qkv + qkv_lora + + qkv_lora = self.fused_qkv_lora(hidden_states, lora_params, + self.layer_idx) + if qkv_lora is not None: + qkv = qkv + qkv_lora + + q, k, v = self.apply_rope(qkv, position_ids) + + out_scale = None + if self.o_proj.has_fp8_qdq or self.o_proj.has_nvfp4 or self.o_proj.has_fp8_block_scales: + out_scale = self.o_proj.inv_input_scale + + q, k, v = self.convert_qkv(q, k, v) + attn_output = self.attn.forward( + q, + k, + v, + attn_metadata, + out_scale=out_scale, + attention_mask=attention_mask, + mrope_config=mrope_config, + attention_window_size=attention_window_size) + hidden_states = attn_output + attn_output = self.o_proj(attn_output, + all_reduce_params=all_reduce_params, + lora_params=lora_params, + layer_idx=self.layer_idx) + return attn_output + + def apply_qk_norm(self, q, k): + raise NotImplementedError( + f"QK norm is not implemented for {self.__class__.__name__}." + "Please override the `apply_qk_norm` method in the subclass.") + + def apply_rope(self, qkv: torch.Tensor, position_ids: torch.Tensor): + """ + Apply RoPE to the query and key, possibly including QK norm. + Args: + qkv (torch.Tensor): The query, key, and value tensor. + position_ids (torch.Tensor): The position IDs of each token for RoPE. + Returns: + tuple: A tuple of (q, k, v). + This method could be overridden in the subclass, it is possible that k/v is None and q is the concatenated qkv tensor, up to the implementation. + Before self.attn.forward, convert_qkv will be called to make sure that the format of (q, k, v) satisfies the requirement of self.attn. + """ + q, k, v = qkv, None, None + if self.qk_norm_type == QkNormType.pre_rope: + q, k, v = self.split_qkv(q, k, v) + q, k = self.apply_qk_norm(q, k) + if not self.enable_rope_fusion and position_ids is not None: + q, k, v = self.split_qkv(q, k, v) + q, k = self.rotary_emb(position_ids, [q, k]) + if self.qk_norm_type == QkNormType.post_rope: + q, k = self.apply_qk_norm(q, k) + + return q, k, v + + +def extract_extra_attrs(layer_idx: str): + extra_attrs = get_model_extra_attrs() + assert extra_attrs is not None, "Model extra attrs is not set" + + metadata_ref = extra_attrs.get("attention_metadata", None) + assert metadata_ref is not None, "Attention metadata is not set" + metadata = metadata_ref() + assert isinstance( + metadata, + TrtllmAttentionMetadata, + ) + + mla_layers = extra_attrs.get("mla_layers", None) + assert mla_layers is not None, "MLA layers is not registered" + mla_layer_ref = mla_layers.get(layer_idx, None) + assert mla_layer_ref is not None, f"Cannot find MLA layer for layer {layer_idx}" + mla_layer = mla_layer_ref() + assert isinstance( + mla_layer, + MLA), "MLA layer must be a subclass of MLA or an instance of MLA" + + return metadata, mla_layer + + +@torch.library.custom_op("trtllm::mla_custom_op", mutates_args=()) +def mla_custom_op( + position_ids: Optional[torch.Tensor], + hidden_states: torch.Tensor, + layer_idx: str, +) -> torch.Tensor: + metadata, mla_layer = extract_extra_attrs(layer_idx) + + return mla_layer.forward_impl(position_ids, hidden_states, metadata) + + +@mla_custom_op.register_fake +def _(position_ids, hidden_states, layer_idx): + _, mla_layer = extract_extra_attrs(layer_idx) + return mla_layer.forward_impl_fake(hidden_states) + + +class MLA(nn.Module): + + def __init__( + self, + *, + hidden_size: int, + num_attention_heads: int, + num_key_value_heads: int, + qk_nope_head_dim: int, + qk_rope_head_dim: int, + v_head_dim: int, + q_lora_rank: int, + kv_lora_rank: int, + predicted_tokens_per_seq: int, + max_position_embeddings: int, + bias: bool, + aux_stream: Optional[torch.cuda.Stream] = None, + pos_embd_params: Optional[PositionalEmbeddingParams] = None, + layer_idx: Optional[int] = None, + dtype: torch.dtype = None, + dense_bias: Optional[bool] = None, + config: Optional[ModelConfig] = None, + ): + """ + Initialize the MLA module. + + Args: + hidden_size (int): The size of the hidden dimension. + num_attention_heads (int): The number of attention heads. + num_key_value_heads (int): The number of key value heads. + qk_nope_head_dim (int): The dimension of the query and key without Rope. + qk_rope_head_dim (int): The dimension of the Rope of query and key. + v_head_dim (int): The dimension of the value. + q_lora_rank (int): The dimension of the compressed query. + kv_lora_rank (int): The dimension of the compressed key and value. + predicted_tokens_per_seq (int): The number of predicted tokens per sequence. + max_position_embeddings (int): The maximum position embeddings. + bias (bool): Whether to use bias in the linear layers. + aux_stream (Optional[torch.cuda.Stream]): The auxiliary CUDA stream for running operations in two parallel streams. + pos_embd_params (PositionalEmbeddingParams): The positional embedding parameters. + layer_idx (int): The layer index. + dtype (torch.dtype): The data type. + dense_bias (bool): Whether to use bias in the output projection layer. + config (ModelConfig): The model configuration. + """ + super().__init__() + self.layer_idx = layer_idx + self.layer_idx_str = str(layer_idx) + self.dtype = dtype + + self.hidden_size = hidden_size + self.num_heads = num_attention_heads + self.num_key_value_heads = num_key_value_heads + self.num_key_value_groups = self.num_heads // self.num_key_value_heads + self.qk_nope_head_dim = qk_nope_head_dim + self.qk_rope_head_dim = qk_rope_head_dim + self.qk_head_dim = qk_nope_head_dim + qk_rope_head_dim + self.v_head_dim = v_head_dim + self.q_lora_rank = q_lora_rank + self.kv_lora_rank = kv_lora_rank + self.predicted_tokens_per_seq = predicted_tokens_per_seq + self.max_position_embeddings = max_position_embeddings + self.pos_embd_params = pos_embd_params + self.dense_bias = dense_bias + if dense_bias is None: + self.dense_bias = bias + + if self.q_lora_rank is None: + self.q_lora_rank = hidden_size + self.is_lite = True + else: + self.is_lite = False + + assert pos_embd_params is not None, "pos_embd_params must be provided in MLA" + + self.register_to_config = False + if config is not None: + if "mla_layers" not in config.extra_attrs: + config.extra_attrs["mla_layers"] = {} + config.extra_attrs["mla_layers"][self.layer_idx_str] = weakref.ref( + self) + self.register_to_config = True + + # tensor parallel + config = config or ModelConfig() + tp_size = config.mapping.tp_size + pp_size = config.mapping.pp_size + if config.mapping.enable_attention_dp: + tp_size = 1 + + mapping = Mapping( + world_size=tp_size * pp_size, + tp_size=tp_size, + pp_size=pp_size, + rank=config.mapping.rank, + gpus_per_node=config.mapping.gpus_per_node, + enable_attention_dp=config.mapping.enable_attention_dp, + ) + + assert self.num_heads % tp_size == 0 + self.num_heads = self.num_heads // tp_size + self.num_key_value_heads = (self.num_key_value_heads + tp_size - + 1) // tp_size + + rms_norm_eps = config.pretrained_config.rms_norm_eps + quant_config = config.get_quant_config() + self.quant_config = quant_config + + if not self.is_lite: + self.fused_a = Linear( + hidden_size, + self.q_lora_rank + self.kv_lora_rank + self.qk_rope_head_dim, + bias=bias, + dtype=dtype, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init, + use_custom_cublas_mm=True) + + self.q_a_layernorm = RMSNorm(hidden_size=self.q_lora_rank, + eps=rms_norm_eps, + dtype=dtype) + + self.q_b_proj = Linear( + self.q_lora_rank, + tp_size * self.num_heads * self.qk_head_dim, + bias=bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.COLUMN, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init) + else: + self.fused_a = Linear( + hidden_size, + self.kv_lora_rank + self.qk_rope_head_dim, + bias=bias, + dtype=dtype, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init, + use_custom_cublas_mm=True) + + self.q_proj = Linear( + self.q_lora_rank, + tp_size * self.num_heads * self.qk_head_dim, + bias=bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.COLUMN, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init, + ) + self.q_b_proj = self.q_proj + + self.kv_a_layernorm = RMSNorm(hidden_size=kv_lora_rank, + dtype=dtype, + eps=rms_norm_eps) + + self.kv_b_proj = Linear( + self.kv_lora_rank, + tp_size * self.num_heads * + (self.qk_nope_head_dim + self.v_head_dim), + bias=bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.COLUMN, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init) + # This parameter will view into self.kv_b_proj.weight after loading weights. + # For dummy weight initialization, this parameter is initialized with empty tensor. + # Used in forward_generation only + self.v_b_proj = nn.Parameter( + torch.empty( + (self.num_heads, self.v_head_dim, self.kv_lora_rank), + dtype=dtype, + ), + requires_grad=False, + ) + + self.o_proj = Linear( + self.num_key_value_heads * self.v_head_dim * tp_size, + self.hidden_size, + bias=self.dense_bias, + dtype=dtype, + mapping=mapping, + tensor_parallel_mode=TensorParallelMode.ROW, + quant_config=quant_config, + skip_create_weights_in_init=config.skip_create_weights_in_init, + ) + + def yarn_get_mscale(scale=1, mscale=1): + if scale <= 1: + return 1.0 + return 0.1 * mscale * math.log(scale) + 1.0 + + mscale_all_dim = pos_embd_params.rope.mscale_all_dim + scaling_factor = pos_embd_params.rope.scale + mscale = yarn_get_mscale(scaling_factor, mscale_all_dim) + q_scaling = 1.0 / (mscale * mscale) + + self.mha = create_attention( + config.attn_backend, + self.layer_idx, + self.num_heads, + head_dim=self.qk_head_dim, + num_kv_heads=self.num_key_value_heads, + pos_embd_params=pos_embd_params, + quant_config=quant_config, + q_scaling=q_scaling, + is_mla_enable=True, + q_lora_rank=self.q_lora_rank, + kv_lora_rank=self.kv_lora_rank, + qk_nope_head_dim=self.qk_nope_head_dim, + qk_rope_head_dim=self.qk_rope_head_dim, + v_head_dim=self.v_head_dim, + predicted_tokens_per_seq=self.predicted_tokens_per_seq, + skip_create_weights_in_init=config.skip_create_weights_in_init, + ) + + self.mqa = create_attention( + config.attn_backend, + self.layer_idx, + self.num_heads, + head_dim=self.kv_lora_rank + self.qk_rope_head_dim, + num_kv_heads=1, + pos_embd_params=pos_embd_params, + quant_config=quant_config, + q_scaling=q_scaling, + is_mla_enable=True, + q_lora_rank=self.q_lora_rank, + kv_lora_rank=self.kv_lora_rank, + qk_nope_head_dim=self.qk_nope_head_dim, + qk_rope_head_dim=self.qk_rope_head_dim, + v_head_dim=self.kv_lora_rank, + predicted_tokens_per_seq=self.predicted_tokens_per_seq, + skip_create_weights_in_init=config.skip_create_weights_in_init, + ) + + self.aux_stream = aux_stream + self.ln_events = [torch.cuda.Event(), torch.cuda.Event()] + + self.enable_rope_fusion = self.mha.support_fused_rope() + self.support_fused_qkv = self.mha.support_fused_qkv() + self.rotary_emb = RotaryEmbedding( + pos_embd_params.rope, + head_dim=self.qk_rope_head_dim, + is_neox=pos_embd_params.is_neox, + ) + self.apply_rotary_emb = not self.enable_rope_fusion + + if not config.skip_create_weights_in_init: + self.create_weights() + + def create_weights(self): + # self.mha/mqa has no weights but has states that are related to quant_config, + # which could be modified after __init__ + self.mha.update_quant_config(self.quant_config) + self.mqa.update_quant_config(self.quant_config) + + # k_b_proj_trans's dtype must be consistent with self.kv_b_proj, + # which can be modified after __init__ + has_fp8_block_scales = ( + self.kv_b_proj.quant_config + and self.kv_b_proj.quant_config.quant_mode.has_fp8_block_scales()) + + mla_weight_dtype = torch.float8_e4m3fn if has_fp8_block_scales else self.dtype + self.k_b_proj_trans = nn.Parameter( + torch.empty( + (self.num_heads, self.kv_lora_rank, self.qk_nope_head_dim), + dtype=mla_weight_dtype, + ), + requires_grad=False, + ) + + if has_fp8_block_scales: + self.k_b_proj_trans_scale = nn.Parameter( + torch.empty( + ( + self.num_heads, + self.kv_lora_rank // 128, + self.qk_nope_head_dim // 128, + ), + dtype=torch.float32, + ), + requires_grad=False, + ) + # This parameter will view into self.kv_b_proj.weight_scale after loading weights. + # For dummy weight initialization, this parameter is initialized with empty tensor. + self.v_b_proj_scale = nn.Parameter( + torch.empty( + ( + self.num_heads, + self.v_head_dim // 128, + self.kv_lora_rank // 128, + ), + dtype=torch.float32, + ), + requires_grad=False, + ) + else: + self.k_b_proj_trans_scale = None + self.v_b_proj_scale = None + + def apply_rope( + self, + q: torch.Tensor, + k_pe: torch.Tensor, + position_ids: torch.Tensor, + ) -> torch.Tensor: + q = q.view(-1, self.num_heads, self.qk_head_dim) + q_pe = q[..., self.qk_nope_head_dim:].reshape( + -1, self.num_heads * self.qk_rope_head_dim) + q_pe, k_pe = self.rotary_emb(position_ids, [q_pe, k_pe]) + q[..., self.qk_nope_head_dim:] = q_pe.view(-1, self.num_heads, + self.qk_rope_head_dim) + return k_pe + + def forward_impl_fake(self, hidden_states: torch.Tensor): + num_tokens = hidden_states.shape[0] + hidden_size = self.o_proj.in_features + return hidden_states.new_empty([num_tokens, hidden_size], + dtype=hidden_states.dtype) + + def forward_impl( + self, + position_ids: Optional[torch.Tensor], + hidden_states: torch.Tensor, + attn_metadata: AttentionMetadata, + ) -> torch.Tensor: + """ + Forward pass for the MLA module. + + Args: + position_ids (Optional[torch.LongTensor]): The position IDs. + hidden_states (torch.Tensor): The hidden states. + attn_metadata (AttentionMetadata): The attention metadata. + all_reduce_params (Optional[AllReduceParams]): The all reduce parameters. + + Returns: + torch.Tensor: The output tensor. + """ + if self.is_lite: + compressed_kv, k_pe = self.fused_a(hidden_states).split( + [self.kv_lora_rank, self.qk_rope_head_dim], -1) + compressed_kv = self.kv_a_layernorm(compressed_kv) + q = hidden_states + else: + q, compressed_kv, k_pe = self.fused_a(hidden_states).split( + [self.q_lora_rank, self.kv_lora_rank, self.qk_rope_head_dim], + -1) + + q, compressed_kv = maybe_execute_in_parallel( + lambda: self.q_a_layernorm(q), + lambda: self.kv_a_layernorm(compressed_kv), + self.ln_events[0], + self.ln_events[1], + self.aux_stream, + ) + + q, latent_cache = maybe_execute_in_parallel( + lambda: self.q_b_proj(q), + lambda: torch.concat([compressed_kv, k_pe], dim=-1), + self.ln_events[0], + self.ln_events[1], + self.aux_stream, + ) + + # split q, k, v into context and gen batches + num_contexts = attn_metadata.num_contexts + num_generations = attn_metadata.num_generations + num_ctx_tokens = attn_metadata.num_ctx_tokens + num_tokens = attn_metadata.num_tokens + + assert q.shape[ + 0] == num_tokens, f"Expect q.shape[0] to be {num_tokens}, but got {q.shape[0]}" + + if num_contexts > 0: + q_ctx = q[:num_ctx_tokens, ...] + compressed_kv_ctx = compressed_kv[:num_ctx_tokens, ...] + k_pe_ctx = k_pe[:num_ctx_tokens, ...] + latent_cache_ctx = latent_cache[:num_ctx_tokens, ...] + if self.apply_rotary_emb: + assert position_ids is not None + k_pe_ctx = self.apply_rope(q_ctx, k_pe_ctx, position_ids) + + attn_output_context = self.forward_context(q_ctx, compressed_kv_ctx, + k_pe_ctx, attn_metadata, + latent_cache_ctx, + position_ids) + else: + attn_output_context = None + + if num_generations > 0: + q_gen = q[num_ctx_tokens:, ...] + compressed_kv_gen = compressed_kv[num_ctx_tokens:, ...] + k_pe_gen = k_pe[num_ctx_tokens:, ...] + latent_cache_gen = latent_cache[num_ctx_tokens:, ...] + if self.apply_rotary_emb: + assert position_ids is not None + k_pe_gen = self.apply_rope(q_gen, k_pe_gen, position_ids) + + attn_output_gen = self.forward_generation(q_gen, compressed_kv_gen, + k_pe_gen, attn_metadata, + latent_cache_gen) + else: + attn_output_gen = None + + # release pytorch activation memory + q = None + compressed_kv = None + k_pe = None + + # merge context and gen batches + if attn_output_context is not None and attn_output_gen is not None: + assert ( + len(attn_output_context.shape) == 2 + ), f"attn_output_context must be rank 2, not {len(attn_output_context.shape)}" + assert ( + len(attn_output_gen.shape) == 2 + ), f"attn_output_gen must be rank 2, not {len(attn_output_gen.shape)}" + attn_output = torch.cat([attn_output_context, attn_output_gen], + dim=0) + # release pytorch activation memory + attn_output_context = None + attn_output_gen = None + elif attn_output_gen is None: + attn_output = attn_output_context + else: + attn_output = attn_output_gen + + return attn_output + + def _maybe_concat_qkv(self, q, k, v): + if k is not None and v is not None and self.support_fused_qkv: + qkv = torch.concat([q, k, v], dim=-1) + q, k, v = qkv, None, None + return q, k, v + + def forward_context_default( + self, + q: torch.Tensor, + compressed_kv: torch.Tensor, + k_pe: torch.Tensor, + attn_metadata: AttentionMetadata, + latent_cache: Optional[torch.Tensor] = None, + ) -> torch.Tensor: + kv = self.kv_b_proj(compressed_kv) + k_nope, v = kv.split( + [ + self.num_heads * self.qk_nope_head_dim, + self.num_heads * self.v_head_dim + ], + -1, + ) + + k = torch.empty_like(q).view(-1, self.num_heads, self.qk_head_dim) + k[..., :self.qk_nope_head_dim] = k_nope.view(-1, self.num_heads, + self.qk_nope_head_dim) + if self.apply_rotary_emb: + k[..., self.qk_nope_head_dim:] = k_pe.view(-1, 1, + self.qk_rope_head_dim) + k = k.view(-1, self.num_heads * self.qk_head_dim) + + # May concat q(including q_pe), k + k_pe, v together + q, k, v = self._maybe_concat_qkv(q, k, v) + + # out_scale = getattr(self.o_proj, "inv_input_scale", None) + out_scale = None # Currently we use BF16 MHA for context phase + + attn_output = self.mha.forward( + q, + k, + v, + attn_metadata, + attention_input_type=AttentionInputType.context_only, + latent_cache=latent_cache, + out_scale=out_scale, + ) + + return attn_output + + def forward_context_with_cached_kv( + self, + q: torch.Tensor, + compressed_kv: torch.Tensor, + k_pe: torch.Tensor, + attn_metadata: AttentionMetadata, + position_ids: Optional[torch.LongTensor] = None, + ) -> torch.Tensor: + trtllm_attention = cast(TrtllmAttention, self.mha) + # split current q into q_nope and q_pe + q_nope, q_pe = q.view([ + -1, self.num_heads, self.qk_nope_head_dim + self.qk_rope_head_dim + ]).split([self.qk_nope_head_dim, self.qk_rope_head_dim], dim=-1) + + # apply rope to current q_pe and k_pe + assert position_ids is not None + assert position_ids.dim() == 1 or (position_ids.dim() == 2 + and position_ids.shape[0] == 1) + assert self.rotary_emb is not None + assert self.rotary_emb.head_dim == self.qk_rope_head_dim + assert q_pe.shape[0] == k_pe.shape[0] + q_pe = q_pe.contiguous().view(-1, + self.num_heads * self.qk_rope_head_dim) + q_pe, k_pe = self.rotary_emb( + position_ids[..., :attn_metadata.num_ctx_tokens], [q_pe, k_pe]) + k_pe = k_pe.contiguous() + + # build q for attention op + q_view = q.view(-1, self.num_heads, + self.qk_nope_head_dim + self.qk_rope_head_dim) + q_view[:, :, + self.qk_nope_head_dim:] = q_pe.view(-1, self.num_heads, + self.qk_rope_head_dim) + q = q_view.view( + -1, + self.num_heads * (self.qk_nope_head_dim + self.qk_rope_head_dim)) + assert q.is_contiguous() + + # append paged kv cache for mla + trtllm_attention.append_paged_kv_cache_for_mla( + compressed_kv, + k_pe, + attn_metadata, + ) + + # copy full_compressed_kv and full_k_pe from paged kv cache + full_compressed_kv, full_k_pe = trtllm_attention.load_paged_kv_cache_for_mla( + attn_metadata, q.dtype) + assert full_compressed_kv.shape[ + 0] == attn_metadata.num_ctx_cached_tokens + attn_metadata.num_ctx_tokens + assert full_compressed_kv.shape[1] == self.kv_lora_rank + assert full_k_pe.shape[ + 0] == attn_metadata.num_ctx_cached_tokens + attn_metadata.num_ctx_tokens + assert full_k_pe.shape[1] == self.qk_rope_head_dim + assert full_compressed_kv.is_contiguous() + assert full_k_pe.is_contiguous() + + # compute full_k_nope and full_v from full_compressed_kv + full_kv = self.kv_b_proj(full_compressed_kv) + full_k_nope, full_v = full_kv.split( + [ + self.num_heads * self.qk_nope_head_dim, + self.num_heads * self.v_head_dim + ], + -1, + ) + full_k_nope = full_k_nope.view(-1, self.num_heads, + self.qk_nope_head_dim) + full_v = full_v.view(-1, self.num_heads, self.v_head_dim) + + # build full_k and full_v + tokens_per_block = attn_metadata.kv_cache_manager.tokens_per_block + # paged kv cache should be initialized to 0 to avoid NaN + paged_full_kv = torch.zeros([ + attn_metadata.num_contexts, 2, + (attn_metadata.max_ctx_kv_len + tokens_per_block - 1) // + tokens_per_block, self.num_heads, tokens_per_block, + max(self.qk_nope_head_dim + self.qk_rope_head_dim, self.v_head_dim) + ], + dtype=q.dtype, + device=q.device) + mla_context_kv_cache_block_offsets = trtllm_attention.set_paged_kv_cache_for_mla( + paged_full_kv, + full_k_nope, + full_v, + full_k_pe, + attn_metadata, + ) + + # out_scale = getattr(self.o_proj, "inv_input_scale", None) + out_scale = None # Currently we use BF16 MHA for context phase + + attn_output = self.mha.forward( + q, + None, + None, + attn_metadata, + attention_input_type=AttentionInputType.context_only, + latent_cache=None, + out_scale=out_scale, + mla_context_paged_kv=paged_full_kv, + mla_context_kv_cache_block_offsets= + mla_context_kv_cache_block_offsets, + ) + + return attn_output + + def forward_context( + self, + q: torch.Tensor, + compressed_kv: torch.Tensor, + k_pe: torch.Tensor, + attn_metadata: AttentionMetadata, + latent_cache: Optional[torch.Tensor] = None, + position_ids: Optional[torch.LongTensor] = None, + ) -> torch.Tensor: + if isinstance(self.mha, TrtllmAttention): + assert isinstance(attn_metadata, TrtllmAttentionMetadata) + trtllm_attention = cast(TrtllmAttention, self.mha) + if trtllm_attention.has_cached_kv_for_mla_context(attn_metadata): + return self.forward_context_with_cached_kv( + q, compressed_kv, k_pe, attn_metadata, position_ids) + return self.forward_context_default(q, compressed_kv, k_pe, + attn_metadata, latent_cache) + + def forward_generation( + self, + q: torch.Tensor, + compressed_kv: torch.Tensor, + k_pe: torch.Tensor, + attn_metadata: AttentionMetadata, + latent_cache: Optional[torch.Tensor] = None, + ) -> torch.Tensor: + num_tokens = q.shape[0] + q_nope, q_pe = q.view([-1, self.num_heads, self.qk_head_dim]).split( + [self.qk_nope_head_dim, self.qk_rope_head_dim], dim=-1) + + # fused_q contains 1) the result of the following bmm with shape [num_tokens, num_heads, kv_lora_rank] + # 2) rope(q_pe) with shape [num_tokens, num_heads, qk_rope_head_dim]. rope is applied inside AttentionOp + fused_q = torch.empty( + [ + num_tokens, self.num_heads, + (self.kv_lora_rank + self.qk_rope_head_dim) + ], + dtype=q.dtype, + device=q.device, + ) + + if self.k_b_proj_trans.dtype == torch.bfloat16: + # [num_heads, num_tokens, self.qk_nope_head_dim] + q_nope_t = q_nope.transpose(0, 1) + # [num_heads, num_tokens, self.kv_lora_rank] + q_nope_out = fused_q[..., :self.kv_lora_rank].transpose(0, 1) + + # [num_heads, num_tokens, self.qk_nope_head_dim] x [num_heads, kv_lora_rank, qk_nope_head_dim] + # -> [num_heads, num_tokens, kv_lora_rank] -> [num_tokens, num_heads, kv_lora_rank] + # The output of bmm is written directly into fused_q + torch.ops.trtllm.bmm_out(q_nope_t, + self.k_b_proj_trans.transpose(1, 2), + q_nope_out) + elif self.k_b_proj_trans.dtype == torch.float8_e4m3fn: + q_nope_fp8, q_nope_scales = torch.ops.trtllm.fp8_batched_quantize_1x128_permute102( + q_nope) + # [num_heads, num_tokens, self.kv_lora_rank] + q_nope_out = fused_q[..., :self.kv_lora_rank].transpose(0, 1) + + torch.ops.trtllm.fp8_block_scaling_bmm_out( + q_nope_fp8, self.k_b_proj_trans, q_nope_scales, + self.k_b_proj_trans_scale, q_nope_out) + q_nope_scales = None + else: + raise NotImplementedError( + f"Missing bmm impl for dtype: {self.k_b_proj_trans.dtype}.") + + if self.apply_rotary_emb: + fused_q[..., self.kv_lora_rank:] = q_pe + fused_q = fused_q.view([ + num_tokens, + self.num_heads * (self.kv_lora_rank + self.qk_rope_head_dim) + ]) + + # out_scale = getattr(self.o_proj, "inv_input_scale", None) + out_scale = None # Although we use FP8 MLA for generation phase, the output is still in BF16 + + attn_out_latent = self.mqa.forward( + fused_q, + None, + None, + attn_metadata, + attention_input_type=AttentionInputType.generation_only, + out_scale=out_scale, + latent_cache=latent_cache, # kvcache and k_pe + q_pe=q_pe, # used by `invokeMLARopeGeneration` + ) + fused_q = None + + assert (attn_out_latent.shape[0] == q.shape[0] and + attn_out_latent.shape[1] == self.num_heads * self.kv_lora_rank) + + # [seq, num_heads, kv_lora_rank] + attn_out_latent = attn_out_latent.view( + [-1, self.num_heads, self.kv_lora_rank]) + + attn_output = torch.empty([num_tokens, self.num_heads, self.v_head_dim], + dtype=attn_out_latent.dtype, + device=attn_out_latent.device) + + if self.v_b_proj.dtype == torch.bfloat16: + # [num_heads, seq, kv_lora_rank] x [num_heads, kv_lora_rank, v_head_dim] + # -> [num_heads, seq, v_head_dim] + torch.ops.trtllm.bmm_out(attn_out_latent.transpose(0, 1), + self.v_b_proj.transpose(1, 2), + attn_output.transpose(0, 1)) + elif self.v_b_proj.dtype == torch.float8_e4m3fn: + attn_out_latent, attn_out_latent_scales = torch.ops.trtllm.fp8_batched_quantize_1x128_permute102( + attn_out_latent) + + torch.ops.trtllm.fp8_block_scaling_bmm_out( + attn_out_latent, self.v_b_proj, attn_out_latent_scales, + self.v_b_proj_scale, attn_output.transpose(0, 1)) + attn_out_latent_scales = None + else: + raise NotImplementedError( + f"Missing bmm impl for dtype: {self.v_b_proj.dtype}.") + + # [seq, num_heads * v_head_dim] + return attn_output.flatten(1, 2) + + def forward( + self, + position_ids: Optional[torch.Tensor], + hidden_states: torch.Tensor, + attn_metadata: AttentionMetadata, + all_reduce_params: Optional[AllReduceParams] = None, + ) -> torch.Tensor: + if self.register_to_config: + attn_output = torch.ops.trtllm.mla_custom_op( + position_ids, hidden_states, self.layer_idx_str) + else: + attn_output = self.forward_impl(position_ids, hidden_states, + attn_metadata) + attn_output = self.o_proj(attn_output, + all_reduce_params=all_reduce_params) + return attn_output diff --git a/latest/_downloads/cba6509356738d5d6b4dcb3b7f52cf39/llm_args.py b/latest/_downloads/cba6509356738d5d6b4dcb3b7f52cf39/llm_args.py index a60766d789..0835058eda 100644 --- a/latest/_downloads/cba6509356738d5d6b4dcb3b7f52cf39/llm_args.py +++ b/latest/_downloads/cba6509356738d5d6b4dcb3b7f52cf39/llm_args.py @@ -1,14 +1,17 @@ import json import math +import os from abc import ABC, abstractmethod -from dataclasses import dataclass, field, fields +from dataclasses import dataclass, field from enum import Enum, EnumMeta from pathlib import Path -from typing import Any, ClassVar, Dict, List, Literal, Optional, Union +from typing import (TYPE_CHECKING, Any, ClassVar, Dict, List, Literal, Optional, + Union) import torch import yaml -from pydantic import BaseModel, Field, validator +from pydantic import (BaseModel, Field, PrivateAttr, field_validator, + model_validator) from strenum import StrEnum from transformers import PreTrainedTokenizerBase @@ -17,23 +20,30 @@ from tensorrt_llm.lora_manager import (LoraConfig, from .._utils import mpi_rank from ..auto_parallel import AutoParallelConfig, infer_cluster_config + +if TYPE_CHECKING: + from tensorrt_llm._torch.pyexecutor.config import PyTorchConfig + # yapf: disable -from ..bindings.executor import BatchingType as _BatchingType -from ..bindings.executor import \ - CacheTransceiverConfig as _CacheTransceiverConfig -from ..bindings.executor import \ - CapacitySchedulerPolicy as _CapacitySchedulerPolicy -from ..bindings.executor import ContextChunkingPolicy as _ContextChunkingPolicy -from ..bindings.executor import DecodingConfig, DecodingMode -from ..bindings.executor import DynamicBatchConfig as _DynamicBatchConfig -from ..bindings.executor import EagleConfig, ExecutorConfig -from ..bindings.executor import \ - ExtendedRuntimePerfKnobConfig as _ExtendedRuntimePerfKnobConfig -from ..bindings.executor import KvCacheConfig as _KvCacheConfig -from ..bindings.executor import \ - LookaheadDecodingConfig as _LookaheadDecodingConfig -from ..bindings.executor import PeftCacheConfig as _PeftCacheConfig -from ..bindings.executor import SchedulerConfig as _SchedulerConfig +# isort: off +from ..bindings.executor import ( + BatchingType as _BatchingType, + CacheTransceiverConfig as _CacheTransceiverConfig, + CapacitySchedulerPolicy as _CapacitySchedulerPolicy, + ContextChunkingPolicy as _ContextChunkingPolicy, + DecodingConfig, + DecodingMode, + DynamicBatchConfig as _DynamicBatchConfig, + EagleConfig as _EagleConfig, + ExecutorConfig as _ExecutorConfig, + ExtendedRuntimePerfKnobConfig as _ExtendedRuntimePerfKnobConfig, + KvCacheConfig as _KvCacheConfig, + LookaheadDecodingConfig as _LookaheadDecodingConfig, + PeftCacheConfig as _PeftCacheConfig, + SchedulerConfig as _SchedulerConfig) # isort: skip +# isort: on +from transformers import PreTrainedTokenizerBase + # yapf: enable from ..builder import BuildConfig, EngineConfig from ..logger import logger @@ -195,7 +205,8 @@ class DecodingBaseConfig(BaseModel): "MTP": MTPDecodingConfig, "Medusa": MedusaDecodingConfig, "Eagle": EagleDecodingConfig, - "Lookahead": LookaheadDecodingConfig + "Lookahead": LookaheadDecodingConfig, + "NGram": NGramDecodingConfig, } config_class = config_classes.get(decoding_type) @@ -228,6 +239,7 @@ class EagleDecodingConfig(DecodingBaseConfig): num_eagle_layers: Optional[int] = None max_non_leaves_per_layer: Optional[int] = None pytorch_eagle_weights_path: Optional[str] = None + eagle3_one_model: Optional[bool] = True @classmethod def from_dict(cls, data: dict): @@ -236,6 +248,40 @@ class EagleDecodingConfig(DecodingBaseConfig): decoding_type: ClassVar[str] = "Eagle" +class NGramDecodingConfig(DecodingBaseConfig): + """ + Configuration for NGram drafter speculative decoding. + + Arguments: + prompt_lookup_num_tokens: int + The length maximum of draft tokens (can be understood as length maximum of output draft tokens). + + max_matching_ngram_size: int + The length maximum of searching tokens (can be understood as length maximum of input tokens to search). + + is_keep_all: bool = True + Whether to keep all candidate pattern-matches pairs, only one match is kept for each pattern if False. + + is_use_oldest: bool = True + Whether to provide the oldest match when pattern is hit, the newest one is provided if False. + + is_public_pool: bool = True + Whether to use a common pool for all requests, or the pool is private for each request if False. + """ + + prompt_lookup_num_tokens: int = 2 + max_matching_ngram_size: int = 4 + is_keep_all: bool = True + is_use_oldest: bool = True + is_public_pool: bool = True + + @classmethod + def from_dict(cls, data: dict): + return cls(**data) + + decoding_type: ClassVar[str] = "NGram" + + class MTPDecodingConfig(DecodingBaseConfig): num_nextn_predict_layers: Optional[int] = 1 use_relaxed_acceptance_for_thinking: Optional[bool] = False @@ -512,7 +558,9 @@ class LookaheadDecodingConfig(DecodingBaseConfig, PybindMirror): get_default_lookahead_decoding_verification_set(), description="Number of NGrams in verification branch per step.") - @validator('max_window_size', 'max_ngram_size', 'max_verification_set_size') + @field_validator('max_window_size', 'max_ngram_size', + 'max_verification_set_size') + @classmethod def validate_positive_values(cls, v): if v <= 0: raise ValueError(f"Value must be positive, got {v}") @@ -699,7 +747,10 @@ class _ModelWrapper: return self.model if isinstance(self.model, str) else None -class LlmArgs(BaseModel): +class BaseLlmArgs(BaseModel): + """ + Base class for both TorchLlmArgs and TrtLlmArgs. It contains all the arguments that are common to both. + """ model_config = { "arbitrary_types_allowed": True, "extra": "allow", @@ -771,20 +822,11 @@ class LlmArgs(BaseModel): cp_config: Optional[dict] = Field(default_factory=dict, description="Context parallel config.") - auto_parallel: bool = Field(default=False, - description="Enable auto parallel mode.") - - auto_parallel_world_size: Optional[int] = Field( - default=None, description="The world size for auto parallel mode.") - load_format: Literal['auto', 'dummy'] = Field( default='auto', description="The format to load the model.", json_schema_extra={"type": "Literal['auto', 'dummy']"}) - enable_tqdm: bool = Field(default=False, - description="Enable tqdm for progress bar.") - # LoRA arguments enable_lora: bool = Field(default=False, description="Enable LoRA.") @@ -816,18 +858,9 @@ class LlmArgs(BaseModel): quant_config: Optional[QuantConfig] = Field( default=None, description="Quantization config.") - calib_config: Optional[CalibConfig] = Field( - default=None, description="Calibration config.") - - # BuildConfig is introduced to give users a familiar interface to configure the model building. - build_config: Optional[object] = Field( - default=None, - description="Build config.", - json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"}) - # Several options from ExecutorConfig, expanded here for less hierarchy - kv_cache_config: Optional[KvCacheConfig] = Field( - default=None, description="KV cache config.") + kv_cache_config: KvCacheConfig = Field(default_factory=KvCacheConfig, + description="KV cache config.") enable_chunked_prefill: bool = Field(default=False, description="Enable chunked prefill.") @@ -850,29 +883,12 @@ class LlmArgs(BaseModel): default=None, description="The maximum number of iterations for request stats.") - workspace: Optional[str] = Field(default=None, - description="The workspace for the model.") - # A handful of options from PretrainedConfig - embedding_parallel_mode: str = Field( - default='SHARDING_ALONG_VOCAB', - description="The embedding parallel mode.") - - fast_build: bool = Field(default=False, description="Enable fast build.") - - # Once set, the model will reuse the build_cache - enable_build_cache: object = Field( - default=False, - description="Enable build cache.", - json_schema_extra={ - "type": f"Union[{get_type_repr(BuildCacheConfig)}, bool]" - }) - peft_cache_config: Optional[PeftCacheConfig] = Field( default=None, description="PEFT cache config.") - scheduler_config: Optional[SchedulerConfig] = Field( - default=None, description="Scheduler config.") + scheduler_config: SchedulerConfig = Field(default_factory=SchedulerConfig, + description="Scheduler config.") cache_transceiver_config: Optional[CacheTransceiverConfig] = Field( default=None, description="Cache transceiver config.") @@ -880,8 +896,8 @@ class LlmArgs(BaseModel): # Speculative decoding parameters speculative_config: Optional[Union[ LookaheadDecodingConfig, MedusaDecodingConfig, EagleDecodingConfig, - MTPDecodingConfig]] = Field(default=None, - description="Speculative decoding config.") + MTPDecodingConfig, NGramDecodingConfig]] = Field( + default=None, description="Speculative decoding config.") batching_type: Optional[BatchingType] = Field(default=None, description="Batching type.") @@ -889,13 +905,6 @@ class LlmArgs(BaseModel): normalize_log_probs: bool = Field( default=False, description="Normalize log probabilities.") - gather_generation_logits: bool = Field( - default=False, description="Gather generation logits.") - - extended_runtime_perf_knob_config: Optional[ - ExtendedRuntimePerfKnobConfig] = Field( - default=None, description="Extended runtime perf knob config.") - max_batch_size: Optional[int] = Field(default=None, description="The maximum batch size.") @@ -916,6 +925,9 @@ class LlmArgs(BaseModel): description="The backend to use.", exclude=True) + gather_generation_logits: bool = Field( + default=False, description="Gather generation logits.") + # private fields those are unstable and just for internal use num_postprocess_workers: int = Field( default=0, @@ -988,40 +1000,19 @@ class LlmArgs(BaseModel): moe_tp_size=self.moe_tensor_parallel_size, moe_ep_size=self.moe_expert_parallel_size, enable_attention_dp=self.enable_attention_dp, - cp_config=self.cp_config, - auto_parallel=self.auto_parallel) - if self.parallel_config.auto_parallel: - self.parallel_config.world_size = self.auto_parallel_world_size - - self.auto_parallel_config = AutoParallelConfig( - sharded_io_allowlist=[ - "past_key_value_\\d+", - "present_key_value_\\d*", - ], - same_buffer_io={ - "past_key_value_(\\d+)": "present_key_value_\\1", - }, - **infer_cluster_config(), - ) - - self.kv_cache_config = self.kv_cache_config or KvCacheConfig() - - self.scheduler_config = self.scheduler_config or SchedulerConfig() - - # This is used to hold th options for convert_checkpoint - self._convert_checkpoint_options = {} + cp_config=self.cp_config) @classmethod - def from_kwargs(cls, **kwargs: Any) -> "LlmArgs": + def from_kwargs(cls, **kwargs: Any) -> "BaseLlmArgs": """Create `LlmArgs` instance from kwargs. Args: kwargs (Any): Arguments passed to `LlmArgs` constructor. Returns: - tensorrt_llm.llmapi.llm_utils.LlmArgs: The `LlmArgs` instance. + tensorrt_llm.llmapi.llm_utils.BaseLlmArgs: The `BaseLlmArgs` instance. """ - kwargs = LlmArgs._maybe_update_config_for_consistency(dict(kwargs)) + kwargs = BaseLlmArgs._maybe_update_config_for_consistency(dict(kwargs)) ret = cls(**kwargs) ret._setup() return ret @@ -1032,8 +1023,7 @@ class LlmArgs(BaseModel): Returns: dict: The dict that contains all fields of the `LlmArgs` instance. """ - return dict( - (field.name, getattr(self, field.name)) for field in fields(self)) + return self.model_dump() @staticmethod def _maybe_update_config_for_consistency( @@ -1041,18 +1031,18 @@ class LlmArgs(BaseModel): # max_beam_width is not included since vague behavior due to lacking the support for dynamic beam width during # generation black_list = set(["max_beam_width"]) - executor_config_attrs = set(attr for attr in dir(ExecutorConfig) - if not attr.startswith('_') - and callable(getattr(ExecutorConfig, attr))) + executor_config_attrs = set( + attr for attr in dir(_ExecutorConfig) if not attr.startswith('_') + and callable(getattr(_ExecutorConfig, attr))) executor_config_attrs -= black_list - llm_args_attr = set(LlmArgs.model_fields.keys()) - # NOTE: When cpp ExecutorConfig add new options, please add the new options into `_LlmArgs` with docs as well + llm_args_attr = set(BaseLlmArgs.model_fields.keys()) + # NOTE: When cpp ExecutorConfig add new options, please add the new options into `LlmArgs` with docs as well # ASK chunweiy for help if you are not sure about the new options. assert executor_config_attrs.issubset( llm_args_attr ), f"New options found in underlying ExecutorConfig: {llm_args_attr - executor_config_attrs}" - # ensure build_config and LlmArgs consistency + # ensure build_config and LlmArgsBase consistency if kwargs_dict.get("backend") != "pytorch" and kwargs_dict.get( "build_config"): # TODO: move this to _perform_config_arbitration() once it's default-on. @@ -1062,11 +1052,11 @@ class LlmArgs(BaseModel): build_val = getattr(kwargs_dict["build_config"], field_name, None) llmargs_val = kwargs_dict.get( - field_name) or LlmArgs.model_fields[field_name] + field_name) or BaseLlmArgs.model_fields[field_name] if build_val != llmargs_val: logger.warning( - f"Overriding LlmArgs.{field_name} ({llmargs_val}) with build_config.{field_name} ({build_val})." + f"Overriding LlmArgsBase.{field_name} ({llmargs_val}) with build_config.{field_name} ({build_val})." ) kwargs_dict[field_name] = build_val @@ -1075,12 +1065,15 @@ class LlmArgs(BaseModel): def _setup(self): ''' This method will setup the configs right before building the model. ''' + is_trt_llm_args = isinstance(self, TrtLlmArgs) + assert isinstance(self.model, (str, Path)), f"Invalid model: {self.model}" - self._setup_embedding_parallel_mode() + if is_trt_llm_args: + self._setup_embedding_parallel_mode() - if self.enable_build_cache: + if is_trt_llm_args and self.enable_build_cache: self.enable_build_cache = BuildCacheConfig() if isinstance( self.enable_build_cache, bool) else self.enable_build_cache if not isinstance(self.enable_build_cache, BuildCacheConfig): @@ -1121,7 +1114,8 @@ class LlmArgs(BaseModel): self.quant_config = self.quant_config or QuantConfig() - self.calib_config = self.calib_config or CalibConfig() + if is_trt_llm_args: + self.calib_config = self.calib_config or CalibConfig() # Note: max_batch_size and max_num_tokens in LlmArgs are for runtime, # which will be passed to the C++ Executor API, overwriting the values @@ -1148,8 +1142,9 @@ class LlmArgs(BaseModel): self.build_config.max_num_tokens = self.max_num_tokens # TODO: remove the checker when manage weights support all data types - if self.fast_build and (self.quant_config.quant_algo is QuantAlgo.FP8 - or self.quant_config.quant_algo is None): + if is_trt_llm_args and self.fast_build and ( + self.quant_config.quant_algo is QuantAlgo.FP8 + or self.quant_config.quant_algo is None): self._update_plugin_config("manage_weights", True) if self.parallel_config._world_size == 1: @@ -1162,9 +1157,12 @@ class LlmArgs(BaseModel): if self.max_lora_rank is not None: self.build_config.lora_config.max_lora_rank = self.max_lora_rank + self._setup_speculative_config() + if self.enable_prompt_adapter: self.build_config.max_prompt_embedding_table_size = self.max_prompt_adapter_token * self.build_config.max_batch_size + def _setup_speculative_config(self): if self.speculative_config: if isinstance(self.speculative_config, LookaheadDecodingConfig): lookahead_config = self.speculative_config @@ -1194,7 +1192,7 @@ class LlmArgs(BaseModel): self.build_config.max_draft_len = self.speculative_config.max_draft_len if self.backend != 'pytorch': - eagle_config = EagleConfig( + eagle_config = _EagleConfig( self.speculative_config.eagle_choices, self.speculative_config.greedy_sampling, self.speculative_config.posterior_threshold, @@ -1207,9 +1205,25 @@ class LlmArgs(BaseModel): from tensorrt_llm._torch.speculative import Eagle3Config self.speculative_config = Eagle3Config( max_draft_tokens=self.speculative_config.max_draft_len, - eagle_weights_path=self.speculative_config. - pytorch_eagle_weights_path) - + draft_model_path=self.speculative_config. + pytorch_eagle_weights_path, + eagle3_one_model=self.speculative_config. + eagle3_one_model) + elif isinstance(self.speculative_config, NGramDecodingConfig): + self.build_config.speculative_decoding_mode = SpeculativeDecodingMode.NGRAM + assert self.backend == 'pytorch' + assert self.speculative_config.prompt_lookup_num_tokens > 0 and self.speculative_config.max_matching_ngram_size > 0 + self.build_config.max_draft_len = self.speculative_config.max_draft_len + from tensorrt_llm._torch.speculative import NGramConfig + self.speculative_config = NGramConfig( + prompt_lookup_num_tokens=self.speculative_config. + prompt_lookup_num_tokens, + max_matching_ngram_size=self.speculative_config. + max_matching_ngram_size, + is_keep_all=self.speculative_config.is_keep_all, + is_use_oldest=self.speculative_config.is_use_oldest, + is_public_pool=self.speculative_config.is_public_pool, + ) elif isinstance(self.speculative_config, MTPDecodingConfig): from tensorrt_llm._torch.speculative import MTPConfig self.speculative_config = MTPConfig( @@ -1350,32 +1364,385 @@ class LlmArgs(BaseModel): f"Invalid embedding_parallel_mode: {self.llm_args.embedding_parallel_mode}" ) - def _validate_kv_cache_config(self): - if self.kv_cache_config is None: - raise ValueError("KvCacheConfig is required for streaming LLM.") - if self.kv_cache_config.max_attention_window is None: - raise ValueError( - "KvCacheConfig.max_attention_window should be set for streaming LLM." - ) - if any(i <= 0 for i in self.kv_cache_config.max_attention_window): - raise ValueError( - "Elements in KvCacheConfig.max_attention_window should be greater than 0." - ) +class TrtLlmArgs(BaseLlmArgs): - if self.kv_cache_config.sink_token_length is None: - raise ValueError( - "KvCacheConfig.sink_token_length should be set for streaming LLM." - ) - if self.kv_cache_config.sink_token_length <= 0: - raise ValueError( - "KvCacheConfig.sink_token_length should be greater than 0.") + auto_parallel: bool = Field( + default=False, + description="Enable auto parallel mode.", + deprecated= + "Use tensor_parallel_size/pipeline_parallel_size/xxx_parallel_size instead.", + ) + auto_parallel_world_size: Optional[int] = Field( + default=None, + description="The world size for auto parallel mode.", + deprecated= + "Use tensor_parallel_size/pipeline_parallel_size/xxx_parallel_size instead.", + ) + + enable_tqdm: bool = Field(default=False, + description="Enable tqdm for progress bar.") + + # BuildConfig is introduced to give users a familiar interface to configure the model building. + build_config: Optional[object] = Field( + default=None, + description="Build config.", + json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"}) + + workspace: Optional[str] = Field(default=None, + description="The workspace for the model.") + + # Once set, the model will reuse the build_cache + enable_build_cache: object = Field( + default=False, + description="Enable build cache.", + json_schema_extra={ + "type": f"Union[{get_type_repr(BuildCacheConfig)}, bool]" + }) + + extended_runtime_perf_knob_config: Optional[ + ExtendedRuntimePerfKnobConfig] = Field( + default=None, description="Extended runtime perf knob config.") + + calib_config: Optional[CalibConfig] = Field( + default=None, description="Calibration config.") + + embedding_parallel_mode: str = Field( + default='SHARDING_ALONG_VOCAB', + description="The embedding parallel mode.") + + fast_build: bool = Field(default=False, description="Enable fast build.") + + # Private attributes + _auto_parallel_config: Optional[AutoParallelConfig] = PrivateAttr( + default=None) + # This is used to hold the options for convert_checkpoint + _convert_checkpoint_options: Dict[str, + Any] = PrivateAttr(default_factory=dict) + + @property + def auto_parallel_config(self) -> AutoParallelConfig: + return self._auto_parallel_config + + @print_traceback_on_error + def model_post_init(self, __context): + super().model_post_init(__context) + + self._auto_parallel_config = AutoParallelConfig( + sharded_io_allowlist=[ + "past_key_value_\\d+", + "present_key_value_\\d*", + ], + same_buffer_io={ + "past_key_value_(\\d+)": "present_key_value_\\1", + }, + **infer_cluster_config(), + ) + + self.parallel_config.auto_parallel = self.auto_parallel + + if self.parallel_config.auto_parallel: + self.parallel_config.world_size = self.auto_parallel_world_size + + +LlmArgs = TrtLlmArgs LLMARGS_EXPLICIT_DOCSTRING = generate_api_docs_as_docstring(LlmArgs, indent=' ' * 4) +class LoadFormat(Enum): + AUTO = 0 + # Initialize all weights randomly. + DUMMY = 1 + + +class TorchLlmArgs(BaseLlmArgs): + + # Just a dummy BuildConfig to allow code reuse with the TrtLlmArgs + build_config: Optional[object] = Field( + default=None, + description="Build config.", + exclude_from_json=True, + json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"}) + + # PyTorch backend specific configurations + + use_cuda_graph: bool = Field( + default=False, + description= + "If true, use CUDA graphs for decoding. CUDA graphs are only created for the batch sizes in cuda_graph_batch_sizes, and are enabled for batches that consist of decoding requests *only* (the reason is that it's hard to capture a single graph with prefill requests since the input shapes are a function of the sequence lengths). Note that each CUDA graph can use up to 200 MB of extra memory." + ) + + cuda_graph_batch_sizes: Optional[List[int]] = Field( + default=None, + description="List of batch sizes to create CUDA graphs for.") + + cuda_graph_max_batch_size: int = Field( + default=0, description="Maximum batch size for CUDA graphs.") + + cuda_graph_padding_enabled: bool = Field( + default=False, + description= + "If true, batches are rounded up to the nearest cuda_graph_batch_size. This is usually a net win for performance." + ) + + disable_overlap_scheduler: bool = Field( + default=False, description="Disable the overlap scheduler.") + + moe_max_num_tokens: Optional[int] = Field( + default=None, + description= + "If set, at most moe_max_num_tokens tokens will be sent to torch.ops.trtllm.fused_moe at the same time. If the number of tokens exceeds moe_max_num_tokens, the input tensors will be split into chunks and a for loop will be used." + ) + + moe_load_balancer: Optional[Union[object, str]] = Field( + default=None, + description="Configuration for MoE load balancing.", + json_schema_extra={"type": "Union[MoeLoadBalancerConfig, str]"}) + + attn_backend: str = Field(default='TRTLLM', + description="Attention backend to use.") + + moe_backend: str = Field(default='CUTLASS', + description="MoE backend to use.") + + mixed_sampler: bool = Field( + default=False, + description= + "If true, will iterate over sampling_params of each request and use the corresponding sampling strategy, e.g. top-k, top-p, etc." + ) + + enable_trtllm_sampler: bool = Field( + default=False, + description= + "If true, will use the TRTLLM sampler instead of the PyTorch sampler. The TRTLLM sampler has a wide coverage of sampling strategies." + ) + + kv_cache_dtype: str = Field(default="auto", + description="Data type for KV cache.") + + use_kv_cache: bool = Field(default=True, + description="Whether to use KV cache.") + + enable_iter_perf_stats: bool = Field( + default=False, description="Enable iteration performance statistics.") + + enable_iter_req_stats: bool = Field( + default=False, + description= + "If true, enables per request stats per iteration. Must also set enable_iter_perf_stats to true to get request stats." + ) + + print_iter_log: bool = Field(default=False, + description="Print iteration logs.") + + torch_compile_enabled: bool = Field( + default=False, description="Enable torch.compile optimization.") + + torch_compile_fullgraph: bool = Field( + default=True, + description="Enable full graph compilation in torch.compile.") + + torch_compile_inductor_enabled: bool = Field( + default=False, description="Enable inductor backend in torch.compile.") + + torch_compile_piecewise_cuda_graph: bool = Field( + default=False, + description="Enable piecewise CUDA graph in torch.compile.") + + torch_compile_enable_userbuffers: bool = Field( + default=True, + description= + "When torch compile is enabled, userbuffers is enabled by default.") + + autotuner_enabled: bool = Field( + default=True, + description="Enable autotuner only when torch compile is enabled.") + + enable_layerwise_nvtx_marker: bool = Field( + default=False, description="If true, enable layerwise nvtx marker.") + + auto_deploy_config: Optional[object] = Field( + default=None, + description="Auto deploy config.", + exclude_from_json=True, + json_schema_extra={"type": f"Optional[AutoDeployConfig]"}) + + load_format: Union[str, LoadFormat] = Field( + default=LoadFormat.AUTO, + description= + "How to load the model weights. By default, detect the weight type from the model checkpoint." + ) + + enable_min_latency: bool = Field( + default=False, + description= + "If true, enable min-latency mode. Currently only used for Llama4.", + ) + + @field_validator('load_format', mode='before') + @classmethod + def convert_load_format(cls, v): + if isinstance(v, LoadFormat): + return v + load_format = v.upper() + if load_format not in LoadFormat.__members__: + raise ValueError(f"Invalid LoadFormat: {v}") + return LoadFormat[load_format] + + # Extra resource managers to use in addition to the KV cache manager. + # Each manager's prepare_resources method is called before the forward pass, + # and update_resources() is called after the pass finishes. free_resources() + # is called when a request finishes. The KV cache manager is guaranteed to + # be invoked after all of these extra managers in all stages. + _extra_resource_managers: Dict[str, + object] = PrivateAttr(default_factory=dict, ) + + @property + def extra_resource_managers(self) -> Dict[str, object]: + return self._extra_resource_managers + + @extra_resource_managers.setter + def extra_resource_managers(self, value: Dict[str, object]) -> None: + self._extra_resource_managers = value + + @print_traceback_on_error + def model_post_init(self, __context): + from .._torch.model_config import MoeLoadBalancerConfig + + super().model_post_init(__context) + self.model_format = _ModelFormatKind.HF + + if isinstance(self.moe_load_balancer, str): + if not os.path.exists(self.moe_load_balancer): + raise FileNotFoundError( + f"MoE load balancer config file not found: {self.moe_load_balancer}" + ) + try: + with open(self.moe_load_balancer) as f: + moe_load_balancer_config = yaml.safe_load(f) + self.moe_load_balancer = MoeLoadBalancerConfig( + **moe_load_balancer_config) + except Exception as e: + raise ValueError( + f"Failed to load MoE load balancer config file: {self.moe_load_balancer}" + ) from e + + # TODO: Remove this after the PyTorch backend is fully migrated to TorchLlmArgs from ExecutorConfig + def get_pytorch_backend_config(self) -> "PyTorchConfig": + from tensorrt_llm._torch.pyexecutor.config import PyTorchConfig + + # TODO: Remove this after the PyTorch backend is fully migrated to TorchLlmArgs from ExecutorConfig + # Just a WAR to support the auto_deploy + if self.auto_deploy_config is not None: + return self.auto_deploy_config + + return PyTorchConfig( + extra_resource_managers=self.extra_resource_managers, + use_cuda_graph=self.use_cuda_graph, + cuda_graph_batch_sizes=self.cuda_graph_batch_sizes, + cuda_graph_max_batch_size=self.cuda_graph_max_batch_size, + cuda_graph_padding_enabled=self.cuda_graph_padding_enabled, + disable_overlap_scheduler=self.disable_overlap_scheduler, + moe_max_num_tokens=self.moe_max_num_tokens, + moe_load_balancer=self.moe_load_balancer, + attn_backend=self.attn_backend, + moe_backend=self.moe_backend, + mixed_sampler=self.mixed_sampler, + enable_trtllm_sampler=self.enable_trtllm_sampler, + kv_cache_dtype=self.kv_cache_dtype, + use_kv_cache=self.use_kv_cache, + enable_iter_perf_stats=self.enable_iter_perf_stats, + enable_iter_req_stats=self.enable_iter_req_stats, + print_iter_log=self.print_iter_log, + torch_compile_enabled=self.torch_compile_enabled, + torch_compile_fullgraph=self.torch_compile_fullgraph, + torch_compile_inductor_enabled=self.torch_compile_inductor_enabled, + torch_compile_piecewise_cuda_graph=self. + torch_compile_piecewise_cuda_graph, + torch_compile_enable_userbuffers=self. + torch_compile_enable_userbuffers, + autotuner_enabled=self.autotuner_enabled, + enable_layerwise_nvtx_marker=self.enable_layerwise_nvtx_marker, + load_format=self.load_format, + enable_min_latency=self.enable_min_latency) + + @field_validator('cuda_graph_max_batch_size') + @classmethod + def validate_cuda_graph_max_batch_size(cls, v): + """Validate cuda_graph_max_batch_size is non-negative.""" + if v < 0: + raise ValueError("cuda_graph_max_batch_size must be non-negative") + return v + + @staticmethod + def _generate_cuda_graph_batch_sizes(max_batch_size: int, + padding_enabled: bool) -> List[int]: + """Generate a list of batch sizes for CUDA graphs. + + Args: + max_batch_size: Maximum batch size to generate up to + padding_enabled: Whether padding is enabled, which affects the batch size distribution + + Returns: + List of batch sizes to create CUDA graphs for + """ + if padding_enabled: + batch_sizes = [1, 2, 4] + [i * 8 for i in range(1, 17)] + else: + batch_sizes = list(range(1, 32)) + [32, 64, 128] + + # Add powers of 2 up to max_batch_size + batch_sizes += [ + 2**i for i in range(8, math.floor(math.log(max_batch_size, 2))) + ] + + # Filter and sort batch sizes + batch_sizes = sorted( + [size for size in batch_sizes if size <= max_batch_size]) + + # Add max_batch_size if not already included + if max_batch_size != batch_sizes[-1]: + batch_sizes.append(max_batch_size) + + return batch_sizes + + @model_validator(mode='after') + def validate_cuda_graph_config(self) -> 'TorchLlmArgs': + """Validate CUDA graph configuration. + + Ensures that: + 1. If cuda_graph_batch_sizes is provided, cuda_graph_max_batch_size must be 0 + 2. If cuda_graph_batch_sizes is not provided, it is generated based on cuda_graph_max_batch_size + 3. If both are provided, cuda_graph_batch_sizes must match the generated values + """ + if self.cuda_graph_batch_sizes is not None: + self.cuda_graph_batch_sizes = sorted(self.cuda_graph_batch_sizes) + if self.cuda_graph_max_batch_size != 0: + if self.cuda_graph_batch_sizes != self._generate_cuda_graph_batch_sizes( + self.cuda_graph_max_batch_size, + self.cuda_graph_padding_enabled): + raise ValueError( + "Please don't set both cuda_graph_batch_sizes " + "and cuda_graph_max_batch_size.\n" + f"cuda_graph_batch_sizes: {self.cuda_graph_batch_sizes}, " + f"cuda_graph_max_batch_size: {self.cuda_graph_max_batch_size}" + ) + else: + self.cuda_graph_max_batch_size = max( + self.cuda_graph_batch_sizes) + else: + max_batch_size = self.cuda_graph_max_batch_size or 128 + generated_sizes = self._generate_cuda_graph_batch_sizes( + max_batch_size, self.cuda_graph_padding_enabled) + self.cuda_graph_batch_sizes = generated_sizes + self.cuda_graph_max_batch_size = max_batch_size + + return self + + def update_llm_args_with_extra_dict( llm_args: Dict, llm_args_dict: Dict, diff --git a/latest/_images/8x_l20_L40S_node_architecture.png b/latest/_images/8x_l20_L40S_node_architecture.png new file mode 100644 index 0000000000..725427f163 Binary files /dev/null and b/latest/_images/8x_l20_L40S_node_architecture.png differ diff --git a/latest/_images/tech_blog3_mla_absorb.png b/latest/_images/tech_blog3_mla_absorb.png new file mode 100644 index 0000000000..ef08436db5 Binary files /dev/null and b/latest/_images/tech_blog3_mla_absorb.png differ diff --git a/latest/_modules/index.html b/latest/_modules/index.html index 7e4551e2ba..d43b59af11 100644 --- a/latest/_modules/index.html +++ b/latest/_modules/index.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -680,6 +684,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/builder.html b/latest/_modules/tensorrt_llm/builder.html index 7c61c14bd0..7d244c86a4 100644 --- a/latest/_modules/tensorrt_llm/builder.html +++ b/latest/_modules/tensorrt_llm/builder.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1986,6 +1990,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/disaggregated_params.html b/latest/_modules/tensorrt_llm/disaggregated_params.html index 76c164667d..d717c7998e 100644 --- a/latest/_modules/tensorrt_llm/disaggregated_params.html +++ b/latest/_modules/tensorrt_llm/disaggregated_params.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -661,6 +665,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/executor/result.html b/latest/_modules/tensorrt_llm/executor/result.html index 4be5107008..5d2f55c78f 100644 --- a/latest/_modules/tensorrt_llm/executor/result.html +++ b/latest/_modules/tensorrt_llm/executor/result.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -760,6 +764,10 @@ output.finish_reason = 'length' elif finish_reasons[src_idx] == tllm.FinishReason.TIMED_OUT: output.finish_reason = 'timeout' + # For disaggregated serving, finish reason might be NOT_FINISHED which is ok + elif finish_reasons[ + src_idx] == tllm.FinishReason.NOT_FINISHED and self.disaggregated_params is not None and self.disaggregated_params.request_type == "context_only": + output.finish_reason = 'not_finished' elif finish_reasons[src_idx] == tllm.FinishReason.CANCELLED: pass else: @@ -1262,6 +1270,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/executor/utils.html b/latest/_modules/tensorrt_llm/executor/utils.html index 641f4b24d3..eeced86c7f 100644 --- a/latest/_modules/tensorrt_llm/executor/utils.html +++ b/latest/_modules/tensorrt_llm/executor/utils.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -510,6 +514,8 @@ from queue import Empty, Queue from typing import Any, Callable, List, NamedTuple, Optional +from strenum import StrEnum + from tensorrt_llm._utils import mpi_rank from tensorrt_llm.bindings.executor import Response from tensorrt_llm.llmapi.utils import print_colored_debug @@ -519,18 +525,35 @@ RemoteMpiCommSessionClient) from ..llmapi.utils import print_colored_debug + +class LlmLauncherEnvs(StrEnum): + # Spawn a process for the LLM-API Proxy + TLLM_SPAWN_PROXY_PROCESS = "TLLM_SPAWN_PROXY_PROCESS" + TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR = "TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR" + TLLM_SPAWN_PROXY_PROCESS_IPC_HMAC_KEY = "TLLM_SPAWN_PROXY_PROCESS_IPC_HMAC_KEY" + + # Whether to use periodical responses handler in await_responses + TLLM_EXECUTOR_PERIODICAL_RESP_IN_AWAIT = "TLLM_EXECUTOR_PERIODICAL_RESP_IN_AWAIT" + + PERIODICAL_RESP_IN_AWAIT = os.getenv( - "TLLM_EXECUTOR_PERIODICAL_RESP_IN_AWAIT") == "1" + LlmLauncherEnvs.TLLM_EXECUTOR_PERIODICAL_RESP_IN_AWAIT) == "1" def get_spawn_proxy_process_ipc_addr_env() -> str | None: ''' Get the IPC address for the spawn proxy process dynamically. ''' - return os.getenv("TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR") + return os.getenv(LlmLauncherEnvs.TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR) + + +def get_spawn_proxy_process_ipc_hmac_key_env() -> bytes | None: + ''' Get the HMAC key for the spawn proxy process dynamically. ''' + if key := os.getenv("TLLM_SPAWN_PROXY_PROCESS_IPC_HMAC_KEY"): + return bytes.fromhex(key) def get_spawn_proxy_process_env() -> bool: ''' Get the environment variable for the spawn proxy process dynamically. ''' - return os.getenv("TLLM_SPAWN_PROXY_PROCESS") == "1" + return os.getenv(LlmLauncherEnvs.TLLM_SPAWN_PROXY_PROCESS) == "1" if PERIODICAL_RESP_IN_AWAIT: @@ -543,14 +566,11 @@ ) == 0, f"create_mpi_comm_session must be called by rank 0, but it was called by rank {mpi_rank()}" if get_spawn_proxy_process_env(): assert get_spawn_proxy_process_ipc_addr_env( - ), "TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR is not set." + ), f"{LlmLauncherEnvs.TLLM_SPAWN_PROXY_PROCESS_IPC_ADDR} is not set." print_colored_debug( f"Using RemoteMpiPoolSessionClient to bind to external MPI processes at {get_spawn_proxy_process_ipc_addr_env()}\n", "yellow") - hmac_key = os.getenv("TLLM_SPAWN_PROXY_PROCESS_IPC_HMAC_KEY") - # Convert the hex string to bytes - if hmac_key is not None: - hmac_key = bytes.fromhex(hmac_key) + hmac_key = get_spawn_proxy_process_ipc_hmac_key_env() return RemoteMpiCommSessionClient( addr=get_spawn_proxy_process_ipc_addr_env(), hmac_key=hmac_key) else: @@ -758,6 +778,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/functional.html b/latest/_modules/tensorrt_llm/functional.html index 90bf679321..e9badfb41b 100644 --- a/latest/_modules/tensorrt_llm/functional.html +++ b/latest/_modules/tensorrt_llm/functional.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -4727,7 +4731,8 @@ UB = 2 AUTO = 3 ONESHOT = 4 - TWOSHOT = 5 + TWOSHOT = 5 + LOWPRECISION = 6 @@ -8673,6 +8678,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/activation.html b/latest/_modules/tensorrt_llm/layers/activation.html index 5ea2653c31..6f42b49a0b 100644 --- a/latest/_modules/tensorrt_llm/layers/activation.html +++ b/latest/_modules/tensorrt_llm/layers/activation.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -639,6 +643,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/attention.html b/latest/_modules/tensorrt_llm/layers/attention.html index a53b5ac23f..d1d370eaf2 100644 --- a/latest/_modules/tensorrt_llm/layers/attention.html +++ b/latest/_modules/tensorrt_llm/layers/attention.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -3504,6 +3508,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/cast.html b/latest/_modules/tensorrt_llm/layers/cast.html index fd8c991724..8a50d31b0e 100644 --- a/latest/_modules/tensorrt_llm/layers/cast.html +++ b/latest/_modules/tensorrt_llm/layers/cast.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -646,6 +650,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/conv.html b/latest/_modules/tensorrt_llm/layers/conv.html index ab22caaf95..83fc9ea691 100644 --- a/latest/_modules/tensorrt_llm/layers/conv.html +++ b/latest/_modules/tensorrt_llm/layers/conv.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -895,6 +899,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/embedding.html b/latest/_modules/tensorrt_llm/layers/embedding.html index 33be0cfd8b..aacfd70035 100644 --- a/latest/_modules/tensorrt_llm/layers/embedding.html +++ b/latest/_modules/tensorrt_llm/layers/embedding.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1362,6 +1366,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/linear.html b/latest/_modules/tensorrt_llm/layers/linear.html index 5cf9f0c957..f399188379 100644 --- a/latest/_modules/tensorrt_llm/layers/linear.html +++ b/latest/_modules/tensorrt_llm/layers/linear.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1210,6 +1214,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/mlp.html b/latest/_modules/tensorrt_llm/layers/mlp.html index 37c99f6445..e5bfd99f21 100644 --- a/latest/_modules/tensorrt_llm/layers/mlp.html +++ b/latest/_modules/tensorrt_llm/layers/mlp.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1236,6 +1240,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/normalization.html b/latest/_modules/tensorrt_llm/layers/normalization.html index 9b65eb238b..39cca5e8ac 100644 --- a/latest/_modules/tensorrt_llm/layers/normalization.html +++ b/latest/_modules/tensorrt_llm/layers/normalization.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1000,6 +1004,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/layers/pooling.html b/latest/_modules/tensorrt_llm/layers/pooling.html index 16ebe7e6a7..3b9b232be7 100644 --- a/latest/_modules/tensorrt_llm/layers/pooling.html +++ b/latest/_modules/tensorrt_llm/layers/pooling.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -655,6 +659,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/llmapi/build_cache.html b/latest/_modules/tensorrt_llm/llmapi/build_cache.html index 8ac8be5d16..211ec0ce6a 100644 --- a/latest/_modules/tensorrt_llm/llmapi/build_cache.html +++ b/latest/_modules/tensorrt_llm/llmapi/build_cache.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -939,6 +943,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/llmapi/llm.html b/latest/_modules/tensorrt_llm/llmapi/llm.html index 230bacb8e0..9f22875735 100644 --- a/latest/_modules/tensorrt_llm/llmapi/llm.html +++ b/latest/_modules/tensorrt_llm/llmapi/llm.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -515,6 +519,7 @@ from tqdm import tqdm from transformers import PreTrainedTokenizerBase +from tensorrt_llm.builder import BuildConfig from tensorrt_llm.inputs.data import TextPrompt from tensorrt_llm.inputs.registry import DefaultInputProcessor @@ -532,8 +537,9 @@ from ..inputs import PromptInputs, create_input_processor, prompt_inputs from ..logger import logger from ..sampling_params import SamplingParams -from .llm_args import LLMARGS_EXPLICIT_DOCSTRING, PybindMirror -from .llm_utils import (CachedModelLoader, KvCacheRetentionConfig, LlmArgs, +from .llm_args import (LLMARGS_EXPLICIT_DOCSTRING, PybindMirror, TorchLlmArgs, + TrtLlmArgs) +from .llm_utils import (CachedModelLoader, KvCacheRetentionConfig, LlmBuildStats, ModelLoader, _ModelRuntimeContext) from .mpi_session import MpiPoolSession, external_mpi_comm_available from .tokenizer import TokenizerBase, _xgrammar_tokenizer_info @@ -625,9 +631,10 @@ self._executor_cls = kwargs.pop("executor_cls", GenerationExecutor) try: - self.pytorch_backend_config = kwargs.pop('pytorch_backend_config', - None) - self.args = LlmArgs.from_kwargs( + llm_args_cls = TorchLlmArgs if kwargs.get( + 'backend', None) == 'pytorch' else TrtLlmArgs + + self.args = llm_args_cls.from_kwargs( model=model, tokenizer=tokenizer, tokenizer_mode=tokenizer_mode, @@ -675,8 +682,9 @@ # Due to the Executor can only accept a engine path, we need to save the engine to a directory self._engine_dir: Optional[Path] = None self._executor: Optional[GenerationExecutor] = None - self._workspace = tempfile.TemporaryDirectory( - suffix="-llm-workspace", dir=self.args.workspace) + if self._on_trt_backend: + self._workspace = tempfile.TemporaryDirectory( + suffix="-llm-workspace", dir=self.args.workspace) self._hf_model_dir: Optional[Path] = None @@ -696,7 +704,7 @@ @property def workspace(self) -> Path: - return Path(self._workspace.name) + return Path(self._workspace.name) if self._on_trt_backend else None
    [docs] @@ -808,10 +816,13 @@ """ sampling_params = self._prepare_sampling_params(sampling_params) - if sampling_params.n > self.args.build_config.max_batch_size: - raise ValueError( - f"SamplingParams.n ({sampling_params.n}) should not exceed max_batch_size ({self.args.build_config.max_batch_size})" - ) + # With pytorch backend, py_executor has logic to handle max_tokens of 1, + # so set to 1 to avoid allocating unnecessary KV cache blocks for single request + # TODO: Also support for trt backend + if (disaggregated_params is not None + and disaggregated_params.request_type == "context_only" + and not self._on_trt_backend): + sampling_params.max_tokens = 1 inputs = prompt_inputs(inputs) @@ -839,8 +850,9 @@ prompt = None query_token_ids = inputs.get("query_token_ids", None) elif "prompt" in inputs: - prompt_token_ids, extra_processed_inputs = self.input_processor( - inputs, sampling_params) + with nvtx_range_debug("input_processor"): + prompt_token_ids, extra_processed_inputs = self.input_processor( + inputs, sampling_params) prompt = inputs['prompt'] if extra_processed_inputs is not None: query_token_ids = extra_processed_inputs.get('query_token_ids') @@ -1025,10 +1037,28 @@ f"The sum of prompt length ({prompt_len/self.args.parallel_config.cp_size}) and query length ({query_len}) max_tokens ({sampling_params.max_tokens}) should not exceed " f"max_seq_len ({build_config.max_seq_len})") - if sampling_params.use_beam_search and sampling_params.n > build_config.max_beam_width: - raise ValueError( - f"sampling_params's n ({sampling_params.n}) should not exceed max_beam_width ({build_config.max_beam_width}) when use_beam_search is True" - ) + if sampling_params.use_beam_search and sampling_params.best_of > build_config.max_beam_width: + if sampling_params.n == sampling_params.best_of: + raise ValueError( + f"sampling_params.n ({sampling_params.n}) cannot exceed max_beam_width ({build_config.max_beam_width}) when use_beam_search is True" + ) + else: + raise ValueError( + f"sampling_params.best_of ({sampling_params.best_of}) cannot exceed max_beam_width ({build_config.max_beam_width}) when use_beam_search is True" + ) + + max_batch_size = self.args.max_batch_size + if max_batch_size is None: + max_batch_size = build_config.max_batch_size + if not sampling_params.use_beam_search and sampling_params.best_of > max_batch_size: + if sampling_params.n == sampling_params.best_of: + raise ValueError( + f"sampling_params.n ({sampling_params.n}) cannot exceed max_batch_size ({max_batch_size}) when use_beam_search is False" + ) + else: + raise ValueError( + f"sampling_params.best_of ({sampling_params.best_of}) cannot exceed max_batch_size ({max_batch_size}) when use_beam_search is False" + ) if sampling_params.prompt_logprobs and not build_config.gather_context_logits: raise ValueError( @@ -1064,11 +1094,19 @@ self.tokenizer) self.tokenizer = self.input_processor.tokenizer - max_batch_size = self.args.max_batch_size or self.args.build_config.max_batch_size - max_num_tokens = self.args.max_num_tokens or self.args.build_config.max_num_tokens - max_seq_len = self.args.max_seq_len or self.args.build_config.max_seq_len + max_batch_size = self.args.max_batch_size + max_num_tokens = self.args.max_num_tokens + max_seq_len = self.args.max_seq_len + + build_config = self.args.build_config if self._on_trt_backend else BuildConfig( + ) + + max_batch_size = max_batch_size or build_config.max_batch_size + max_num_tokens = max_num_tokens or build_config.max_num_tokens + max_seq_len = max_seq_len or build_config.max_seq_len + executor_config = tllm.ExecutorConfig( - max_beam_width=self.args.build_config.max_beam_width, + max_beam_width=self.args.max_beam_width, scheduler_config=PybindMirror.maybe_to_pybind( self.args.scheduler_config), batching_type=PybindMirror.maybe_to_pybind(self.args.batching_type) @@ -1094,7 +1132,7 @@ if self.args.peft_cache_config is not None: executor_config.peft_cache_config = PybindMirror.maybe_to_pybind( self.args.peft_cache_config) - elif self.args.build_config.plugin_config.lora_plugin: + elif self._on_trt_backend and self.args.build_config.plugin_config.lora_plugin: engine_config = EngineConfig.from_json_file(self._engine_dir / "config.json") lora_config = engine_config.build_config.lora_config @@ -1122,7 +1160,7 @@ executor_config.normalize_log_probs = self.args.normalize_log_probs executor_config.enable_chunked_context = self.args.enable_chunked_prefill executor_config.max_beam_width = self.args.max_beam_width or self.args.build_config.max_beam_width - if self.args.extended_runtime_perf_knob_config is not None: + if self._on_trt_backend and self.args.extended_runtime_perf_knob_config is not None: executor_config.extended_runtime_perf_knob_config = PybindMirror.maybe_to_pybind( self.args.extended_runtime_perf_knob_config) if self.args.cache_transceiver_config is not None: @@ -1132,9 +1170,11 @@ update_executor_config( executor_config, backend=self.args.backend, - pytorch_backend_config=self.pytorch_backend_config, + pytorch_backend_config=self.args.get_pytorch_backend_config() + if self.args.backend == "pytorch" else None, mapping=self.args.parallel_config.to_mapping(), - build_config=self.args.build_config, + build_config=self.args.build_config + if self._on_trt_backend else None, speculative_config=self.args.speculative_config, hf_model_dir=self._hf_model_dir, trt_engine_dir=self._engine_dir, @@ -1142,8 +1182,9 @@ max_seq_len=max_seq_len) executor_config.llm_parallel_config = self.args.parallel_config return_logits = self.args.gather_generation_logits or ( - self.args.build_config + self._on_trt_backend and self.args.build_config and self.args.build_config.gather_context_logits) + self._executor = self._executor_cls.create( self._engine_dir, executor_config=executor_config, @@ -1160,6 +1201,10 @@ is_llm_executor=True, lora_config=self.args.lora_config) + @property + def _on_trt_backend(self) -> bool: + return isinstance(self.args, TrtLlmArgs) + def _try_load_tokenizer(self) -> Optional[TokenizerBase]: if self.args.skip_tokenizer_init: return None @@ -1379,6 +1424,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/llmapi/llm_args.html b/latest/_modules/tensorrt_llm/llmapi/llm_args.html index 5109ab9e3e..3e41a4e02b 100644 --- a/latest/_modules/tensorrt_llm/llmapi/llm_args.html +++ b/latest/_modules/tensorrt_llm/llmapi/llm_args.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -505,15 +509,18 @@

    Source code for tensorrt_llm.llmapi.llm_args

     import json
 import math
+import os
 from abc import ABC, abstractmethod
-from dataclasses import dataclass, field, fields
+from dataclasses import dataclass, field
 from enum import Enum, EnumMeta
 from pathlib import Path
-from typing import Any, ClassVar, Dict, List, Literal, Optional, Union
+from typing import (TYPE_CHECKING, Any, ClassVar, Dict, List, Literal, Optional,
+                    Union)
 
 import torch
 import yaml
-from pydantic import BaseModel, Field, validator
+from pydantic import (BaseModel, Field, PrivateAttr, field_validator,
+                      model_validator)
 from strenum import StrEnum
 from transformers import PreTrainedTokenizerBase
 
@@ -522,23 +529,30 @@
 
 from .._utils import mpi_rank
 from ..auto_parallel import AutoParallelConfig, infer_cluster_config
+
+if TYPE_CHECKING:
+    from tensorrt_llm._torch.pyexecutor.config import PyTorchConfig
+
 # yapf: disable
-from ..bindings.executor import BatchingType as _BatchingType
-from ..bindings.executor import \
-    CacheTransceiverConfig as _CacheTransceiverConfig
-from ..bindings.executor import \
-    CapacitySchedulerPolicy as _CapacitySchedulerPolicy
-from ..bindings.executor import ContextChunkingPolicy as _ContextChunkingPolicy
-from ..bindings.executor import DecodingConfig, DecodingMode
-from ..bindings.executor import DynamicBatchConfig as _DynamicBatchConfig
-from ..bindings.executor import EagleConfig, ExecutorConfig
-from ..bindings.executor import \
-    ExtendedRuntimePerfKnobConfig as _ExtendedRuntimePerfKnobConfig
-from ..bindings.executor import KvCacheConfig as _KvCacheConfig
-from ..bindings.executor import \
-    LookaheadDecodingConfig as _LookaheadDecodingConfig
-from ..bindings.executor import PeftCacheConfig as _PeftCacheConfig
-from ..bindings.executor import SchedulerConfig as _SchedulerConfig
+# isort: off
+from ..bindings.executor import (
+                                 BatchingType as _BatchingType,
+                                 CacheTransceiverConfig as _CacheTransceiverConfig,
+                                 CapacitySchedulerPolicy as _CapacitySchedulerPolicy,
+                                 ContextChunkingPolicy as _ContextChunkingPolicy,
+                                 DecodingConfig,
+                                 DecodingMode,
+                                 DynamicBatchConfig as _DynamicBatchConfig,
+                                 EagleConfig as _EagleConfig,
+                                 ExecutorConfig as _ExecutorConfig,
+                                 ExtendedRuntimePerfKnobConfig as _ExtendedRuntimePerfKnobConfig,
+                                 KvCacheConfig as _KvCacheConfig,
+                                 LookaheadDecodingConfig as _LookaheadDecodingConfig,
+                                 PeftCacheConfig as _PeftCacheConfig,
+                                 SchedulerConfig as _SchedulerConfig) # isort: skip
+# isort: on
+from transformers import PreTrainedTokenizerBase
+
 # yapf: enable
 from ..builder import BuildConfig, EngineConfig
 from ..logger import logger
@@ -709,7 +723,8 @@
             "MTP": MTPDecodingConfig,
             "Medusa": MedusaDecodingConfig,
             "Eagle": EagleDecodingConfig,
-            "Lookahead": LookaheadDecodingConfig
+            "Lookahead": LookaheadDecodingConfig,
+            "NGram": NGramDecodingConfig,
         }
 
         config_class = config_classes.get(decoding_type)
@@ -750,6 +765,7 @@
     num_eagle_layers: Optional[int] = None
     max_non_leaves_per_layer: Optional[int] = None
     pytorch_eagle_weights_path: Optional[str] = None
+    eagle3_one_model: Optional[bool] = True
 
 
    
 [docs]
@@ -762,6 +778,46 @@
 
 
 
+
    
+[docs]
+class NGramDecodingConfig(DecodingBaseConfig):
+    """
+    Configuration for NGram drafter speculative decoding.
+
+    Arguments:
+        prompt_lookup_num_tokens: int
+                The length maximum of draft tokens (can be understood as length maximum of output draft tokens).
+
+        max_matching_ngram_size: int
+            The length maximum of searching tokens (can be understood as length maximum of input tokens to search).
+
+        is_keep_all: bool = True
+            Whether to keep all candidate pattern-matches pairs, only one match is kept for each pattern if False.
+
+        is_use_oldest: bool = True
+            Whether to provide the oldest match when pattern is hit, the newest one is provided if False.
+
+        is_public_pool: bool = True
+            Whether to use a common pool for all requests, or the pool is private for each request if False.
+    """
+
+    prompt_lookup_num_tokens: int = 2
+    max_matching_ngram_size: int = 4
+    is_keep_all: bool = True
+    is_use_oldest: bool = True
+    is_public_pool: bool = True
+
+
    
+[docs]
+    @classmethod
+    def from_dict(cls, data: dict):
+        return cls(**data)
    
+
+
+    decoding_type: ClassVar[str] = "NGram"
    
+
+
+
 
    
 [docs]
 class MTPDecodingConfig(DecodingBaseConfig):
@@ -1063,7 +1119,9 @@
 
 
    
 [docs]
-    @validator('max_window_size', 'max_ngram_size', 'max_verification_set_size')
+    @field_validator('max_window_size', 'max_ngram_size',
+                     'max_verification_set_size')
+    @classmethod
     def validate_positive_values(cls, v):
         if v <= 0:
             raise ValueError(f"Value must be positive, got {v}")
@@ -1270,7 +1328,10 @@
         return self.model if isinstance(self.model, str) else None
 
 
-class LlmArgs(BaseModel):
+class BaseLlmArgs(BaseModel):
+    """
+    Base class for both TorchLlmArgs and TrtLlmArgs. It contains all the arguments that are common to both.
+    """
     model_config = {
         "arbitrary_types_allowed": True,
         "extra": "allow",
@@ -1342,20 +1403,11 @@
     cp_config: Optional[dict] = Field(default_factory=dict,
                                       description="Context parallel config.")
 
-    auto_parallel: bool = Field(default=False,
-                                description="Enable auto parallel mode.")
-
-    auto_parallel_world_size: Optional[int] = Field(
-        default=None, description="The world size for auto parallel mode.")
-
     load_format: Literal['auto', 'dummy'] = Field(
         default='auto',
         description="The format to load the model.",
         json_schema_extra={"type": "Literal['auto', 'dummy']"})
 
-    enable_tqdm: bool = Field(default=False,
-                              description="Enable tqdm for progress bar.")
-
     # LoRA arguments
     enable_lora: bool = Field(default=False, description="Enable LoRA.")
 
@@ -1387,18 +1439,9 @@
     quant_config: Optional[QuantConfig] = Field(
         default=None, description="Quantization config.")
 
-    calib_config: Optional[CalibConfig] = Field(
-        default=None, description="Calibration config.")
-
-    # BuildConfig is introduced to give users a familiar interface to configure the model building.
-    build_config: Optional[object] = Field(
-        default=None,
-        description="Build config.",
-        json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"})
-
     # Several options from ExecutorConfig, expanded here for less hierarchy
-    kv_cache_config: Optional[KvCacheConfig] = Field(
-        default=None, description="KV cache config.")
+    kv_cache_config: KvCacheConfig = Field(default_factory=KvCacheConfig,
+                                           description="KV cache config.")
 
     enable_chunked_prefill: bool = Field(default=False,
                                          description="Enable chunked prefill.")
@@ -1421,29 +1464,12 @@
         default=None,
         description="The maximum number of iterations for request stats.")
 
-    workspace: Optional[str] = Field(default=None,
-                                     description="The workspace for the model.")
-
     # A handful of options from PretrainedConfig
-    embedding_parallel_mode: str = Field(
-        default='SHARDING_ALONG_VOCAB',
-        description="The embedding parallel mode.")
-
-    fast_build: bool = Field(default=False, description="Enable fast build.")
-
-    # Once set, the model will reuse the build_cache
-    enable_build_cache: object = Field(
-        default=False,
-        description="Enable build cache.",
-        json_schema_extra={
-            "type": f"Union[{get_type_repr(BuildCacheConfig)}, bool]"
-        })
-
     peft_cache_config: Optional[PeftCacheConfig] = Field(
         default=None, description="PEFT cache config.")
 
-    scheduler_config: Optional[SchedulerConfig] = Field(
-        default=None, description="Scheduler config.")
+    scheduler_config: SchedulerConfig = Field(default_factory=SchedulerConfig,
+                                              description="Scheduler config.")
 
     cache_transceiver_config: Optional[CacheTransceiverConfig] = Field(
         default=None, description="Cache transceiver config.")
@@ -1451,8 +1477,8 @@
     # Speculative decoding parameters
     speculative_config: Optional[Union[
         LookaheadDecodingConfig, MedusaDecodingConfig, EagleDecodingConfig,
-        MTPDecodingConfig]] = Field(default=None,
-                                    description="Speculative decoding config.")
+        MTPDecodingConfig, NGramDecodingConfig]] = Field(
+            default=None, description="Speculative decoding config.")
 
     batching_type: Optional[BatchingType] = Field(default=None,
                                                   description="Batching type.")
@@ -1460,13 +1486,6 @@
     normalize_log_probs: bool = Field(
         default=False, description="Normalize log probabilities.")
 
-    gather_generation_logits: bool = Field(
-        default=False, description="Gather generation logits.")
-
-    extended_runtime_perf_knob_config: Optional[
-        ExtendedRuntimePerfKnobConfig] = Field(
-            default=None, description="Extended runtime perf knob config.")
-
     max_batch_size: Optional[int] = Field(default=None,
                                           description="The maximum batch size.")
 
@@ -1487,6 +1506,9 @@
                                    description="The backend to use.",
                                    exclude=True)
 
+    gather_generation_logits: bool = Field(
+        default=False, description="Gather generation logits.")
+
     # private fields those are unstable and just for internal use
     num_postprocess_workers: int = Field(
         default=0,
@@ -1559,40 +1581,19 @@
             moe_tp_size=self.moe_tensor_parallel_size,
             moe_ep_size=self.moe_expert_parallel_size,
             enable_attention_dp=self.enable_attention_dp,
-            cp_config=self.cp_config,
-            auto_parallel=self.auto_parallel)
-        if self.parallel_config.auto_parallel:
-            self.parallel_config.world_size = self.auto_parallel_world_size
-
-        self.auto_parallel_config = AutoParallelConfig(
-            sharded_io_allowlist=[
-                "past_key_value_\\d+",
-                "present_key_value_\\d*",
-            ],
-            same_buffer_io={
-                "past_key_value_(\\d+)": "present_key_value_\\1",
-            },
-            **infer_cluster_config(),
-        )
-
-        self.kv_cache_config = self.kv_cache_config or KvCacheConfig()
-
-        self.scheduler_config = self.scheduler_config or SchedulerConfig()
-
-        # This is used to hold th options for convert_checkpoint
-        self._convert_checkpoint_options = {}
+            cp_config=self.cp_config)
 
     @classmethod
-    def from_kwargs(cls, **kwargs: Any) -> "LlmArgs":
+    def from_kwargs(cls, **kwargs: Any) -> "BaseLlmArgs":
         """Create `LlmArgs` instance from kwargs.
 
         Args:
             kwargs (Any): Arguments passed to `LlmArgs` constructor.
 
         Returns:
-            tensorrt_llm.llmapi.llm_utils.LlmArgs: The `LlmArgs` instance.
+            tensorrt_llm.llmapi.llm_utils.BaseLlmArgs: The `BaseLlmArgs` instance.
         """
-        kwargs = LlmArgs._maybe_update_config_for_consistency(dict(kwargs))
+        kwargs = BaseLlmArgs._maybe_update_config_for_consistency(dict(kwargs))
         ret = cls(**kwargs)
         ret._setup()
         return ret
@@ -1603,8 +1604,7 @@
         Returns:
             dict: The dict that contains all fields of the `LlmArgs` instance.
         """
-        return dict(
-            (field.name, getattr(self, field.name)) for field in fields(self))
+        return self.model_dump()
 
     @staticmethod
     def _maybe_update_config_for_consistency(
@@ -1612,18 +1612,18 @@
         # max_beam_width is not included since vague behavior due to lacking the support for dynamic beam width during
         # generation
         black_list = set(["max_beam_width"])
-        executor_config_attrs = set(attr for attr in dir(ExecutorConfig)
-                                    if not attr.startswith('_')
-                                    and callable(getattr(ExecutorConfig, attr)))
+        executor_config_attrs = set(
+            attr for attr in dir(_ExecutorConfig) if not attr.startswith('_')
+            and callable(getattr(_ExecutorConfig, attr)))
         executor_config_attrs -= black_list
-        llm_args_attr = set(LlmArgs.model_fields.keys())
-        # NOTE: When cpp ExecutorConfig add new options, please add the new options into `_LlmArgs` with docs as well
+        llm_args_attr = set(BaseLlmArgs.model_fields.keys())
+        # NOTE: When cpp ExecutorConfig add new options, please add the new options into `LlmArgs` with docs as well
         # ASK chunweiy for help if you are not sure about the new options.
         assert executor_config_attrs.issubset(
             llm_args_attr
         ), f"New options found in underlying ExecutorConfig: {llm_args_attr - executor_config_attrs}"
 
-        # ensure build_config and LlmArgs consistency
+        # ensure build_config and LlmArgsBase consistency
         if kwargs_dict.get("backend") != "pytorch" and kwargs_dict.get(
                 "build_config"):
             # TODO: move this to _perform_config_arbitration() once it's default-on.
@@ -1633,11 +1633,11 @@
                 build_val = getattr(kwargs_dict["build_config"], field_name,
                                     None)
                 llmargs_val = kwargs_dict.get(
-                    field_name) or LlmArgs.model_fields[field_name]
+                    field_name) or BaseLlmArgs.model_fields[field_name]
 
                 if build_val != llmargs_val:
                     logger.warning(
-                        f"Overriding LlmArgs.{field_name} ({llmargs_val}) with build_config.{field_name} ({build_val})."
+                        f"Overriding LlmArgsBase.{field_name} ({llmargs_val}) with build_config.{field_name} ({build_val})."
                     )
                     kwargs_dict[field_name] = build_val
 
@@ -1646,12 +1646,15 @@
     def _setup(self):
         ''' This method will setup the configs right before building the model. '''
 
+        is_trt_llm_args = isinstance(self, TrtLlmArgs)
+
         assert isinstance(self.model,
                           (str, Path)), f"Invalid model: {self.model}"
 
-        self._setup_embedding_parallel_mode()
+        if is_trt_llm_args:
+            self._setup_embedding_parallel_mode()
 
-        if self.enable_build_cache:
+        if is_trt_llm_args and self.enable_build_cache:
             self.enable_build_cache = BuildCacheConfig() if isinstance(
                 self.enable_build_cache, bool) else self.enable_build_cache
             if not isinstance(self.enable_build_cache, BuildCacheConfig):
@@ -1692,7 +1695,8 @@
 
         self.quant_config = self.quant_config or QuantConfig()
 
-        self.calib_config = self.calib_config or CalibConfig()
+        if is_trt_llm_args:
+            self.calib_config = self.calib_config or CalibConfig()
 
         # Note: max_batch_size and max_num_tokens in LlmArgs are for runtime,
         # which will be passed to the C++ Executor API, overwriting the values
@@ -1719,8 +1723,9 @@
                 self.build_config.max_num_tokens = self.max_num_tokens
 
         # TODO: remove the checker when manage weights support all data types
-        if self.fast_build and (self.quant_config.quant_algo is QuantAlgo.FP8
-                                or self.quant_config.quant_algo is None):
+        if is_trt_llm_args and self.fast_build and (
+                self.quant_config.quant_algo is QuantAlgo.FP8
+                or self.quant_config.quant_algo is None):
             self._update_plugin_config("manage_weights", True)
 
         if self.parallel_config._world_size == 1:
@@ -1733,9 +1738,12 @@
             if self.max_lora_rank is not None:
                 self.build_config.lora_config.max_lora_rank = self.max_lora_rank
 
+        self._setup_speculative_config()
+
         if self.enable_prompt_adapter:
             self.build_config.max_prompt_embedding_table_size = self.max_prompt_adapter_token * self.build_config.max_batch_size
 
+    def _setup_speculative_config(self):
         if self.speculative_config:
             if isinstance(self.speculative_config, LookaheadDecodingConfig):
                 lookahead_config = self.speculative_config
@@ -1765,7 +1773,7 @@
                 self.build_config.max_draft_len = self.speculative_config.max_draft_len
 
                 if self.backend != 'pytorch':
-                    eagle_config = EagleConfig(
+                    eagle_config = _EagleConfig(
                         self.speculative_config.eagle_choices,
                         self.speculative_config.greedy_sampling,
                         self.speculative_config.posterior_threshold,
@@ -1778,9 +1786,25 @@
                     from tensorrt_llm._torch.speculative import Eagle3Config
                     self.speculative_config = Eagle3Config(
                         max_draft_tokens=self.speculative_config.max_draft_len,
-                        eagle_weights_path=self.speculative_config.
-                        pytorch_eagle_weights_path)
-
+                        draft_model_path=self.speculative_config.
+                        pytorch_eagle_weights_path,
+                        eagle3_one_model=self.speculative_config.
+                        eagle3_one_model)
+            elif isinstance(self.speculative_config, NGramDecodingConfig):
+                self.build_config.speculative_decoding_mode = SpeculativeDecodingMode.NGRAM
+                assert self.backend == 'pytorch'
+                assert self.speculative_config.prompt_lookup_num_tokens > 0 and self.speculative_config.max_matching_ngram_size > 0
+                self.build_config.max_draft_len = self.speculative_config.max_draft_len
+                from tensorrt_llm._torch.speculative import NGramConfig
+                self.speculative_config = NGramConfig(
+                    prompt_lookup_num_tokens=self.speculative_config.
+                    prompt_lookup_num_tokens,
+                    max_matching_ngram_size=self.speculative_config.
+                    max_matching_ngram_size,
+                    is_keep_all=self.speculative_config.is_keep_all,
+                    is_use_oldest=self.speculative_config.is_use_oldest,
+                    is_public_pool=self.speculative_config.is_public_pool,
+                )
             elif isinstance(self.speculative_config, MTPDecodingConfig):
                 from tensorrt_llm._torch.speculative import MTPConfig
                 self.speculative_config = MTPConfig(
@@ -1921,32 +1945,409 @@
                 f"Invalid embedding_parallel_mode: {self.llm_args.embedding_parallel_mode}"
             )
 
-    def _validate_kv_cache_config(self):
-        if self.kv_cache_config is None:
-            raise ValueError("KvCacheConfig is required for streaming LLM.")
 
-        if self.kv_cache_config.max_attention_window is None:
-            raise ValueError(
-                "KvCacheConfig.max_attention_window should be set for streaming LLM."
-            )
-        if any(i <= 0 for i in self.kv_cache_config.max_attention_window):
-            raise ValueError(
-                "Elements in KvCacheConfig.max_attention_window should be greater than 0."
-            )
+
    
+[docs]
+class TrtLlmArgs(BaseLlmArgs):
 
-        if self.kv_cache_config.sink_token_length is None:
-            raise ValueError(
-                "KvCacheConfig.sink_token_length should be set for streaming LLM."
-            )
-        if self.kv_cache_config.sink_token_length <= 0:
-            raise ValueError(
-                "KvCacheConfig.sink_token_length should be greater than 0.")
+    auto_parallel: bool = Field(
+        default=False,
+        description="Enable auto parallel mode.",
+        deprecated=
+        "Use tensor_parallel_size/pipeline_parallel_size/xxx_parallel_size instead.",
+    )
 
+    auto_parallel_world_size: Optional[int] = Field(
+        default=None,
+        description="The world size for auto parallel mode.",
+        deprecated=
+        "Use tensor_parallel_size/pipeline_parallel_size/xxx_parallel_size instead.",
+    )
+
+    enable_tqdm: bool = Field(default=False,
+                              description="Enable tqdm for progress bar.")
+
+    # BuildConfig is introduced to give users a familiar interface to configure the model building.
+    build_config: Optional[object] = Field(
+        default=None,
+        description="Build config.",
+        json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"})
+
+    workspace: Optional[str] = Field(default=None,
+                                     description="The workspace for the model.")
+
+    # Once set, the model will reuse the build_cache
+    enable_build_cache: object = Field(
+        default=False,
+        description="Enable build cache.",
+        json_schema_extra={
+            "type": f"Union[{get_type_repr(BuildCacheConfig)}, bool]"
+        })
+
+    extended_runtime_perf_knob_config: Optional[
+        ExtendedRuntimePerfKnobConfig] = Field(
+            default=None, description="Extended runtime perf knob config.")
+
+    calib_config: Optional[CalibConfig] = Field(
+        default=None, description="Calibration config.")
+
+    embedding_parallel_mode: str = Field(
+        default='SHARDING_ALONG_VOCAB',
+        description="The embedding parallel mode.")
+
+    fast_build: bool = Field(default=False, description="Enable fast build.")
+
+    # Private attributes
+    _auto_parallel_config: Optional[AutoParallelConfig] = PrivateAttr(
+        default=None)
+    # This is used to hold the options for convert_checkpoint
+    _convert_checkpoint_options: Dict[str,
+                                      Any] = PrivateAttr(default_factory=dict)
+
+    @property
+    def auto_parallel_config(self) -> AutoParallelConfig:
+        return self._auto_parallel_config
+
+
    
+[docs]
+    @print_traceback_on_error
+    def model_post_init(self, __context):
+        super().model_post_init(__context)
+
+        self._auto_parallel_config = AutoParallelConfig(
+            sharded_io_allowlist=[
+                "past_key_value_\\d+",
+                "present_key_value_\\d*",
+            ],
+            same_buffer_io={
+                "past_key_value_(\\d+)": "present_key_value_\\1",
+            },
+            **infer_cluster_config(),
+        )
+
+        self.parallel_config.auto_parallel = self.auto_parallel
+
+        if self.parallel_config.auto_parallel:
+            self.parallel_config.world_size = self.auto_parallel_world_size
    
+
    
+
+
+
+LlmArgs = TrtLlmArgs
 
 LLMARGS_EXPLICIT_DOCSTRING = generate_api_docs_as_docstring(LlmArgs,
                                                             indent=' ' * 4)
 
 
+class LoadFormat(Enum):
+    AUTO = 0
+    # Initialize all weights randomly.
+    DUMMY = 1
+
+
+
    
+[docs]
+class TorchLlmArgs(BaseLlmArgs):
+
+    # Just a dummy BuildConfig to allow code reuse with the TrtLlmArgs
+    build_config: Optional[object] = Field(
+        default=None,
+        description="Build config.",
+        exclude_from_json=True,
+        json_schema_extra={"type": f"Optional[{get_type_repr(BuildConfig)}]"})
+
+    # PyTorch backend specific configurations
+
+    use_cuda_graph: bool = Field(
+        default=False,
+        description=
+        "If true, use CUDA graphs for decoding. CUDA graphs are only created for the batch sizes in cuda_graph_batch_sizes, and are enabled for batches that consist of decoding requests *only* (the reason is that it's hard to capture a single graph with prefill requests since the input shapes are a function of the sequence lengths). Note that each CUDA graph can use up to 200 MB of extra memory."
+    )
+
+    cuda_graph_batch_sizes: Optional[List[int]] = Field(
+        default=None,
+        description="List of batch sizes to create CUDA graphs for.")
+
+    cuda_graph_max_batch_size: int = Field(
+        default=0, description="Maximum batch size for CUDA graphs.")
+
+    cuda_graph_padding_enabled: bool = Field(
+        default=False,
+        description=
+        "If true, batches are rounded up to the nearest cuda_graph_batch_size. This is usually a net win for performance."
+    )
+
+    disable_overlap_scheduler: bool = Field(
+        default=False, description="Disable the overlap scheduler.")
+
+    moe_max_num_tokens: Optional[int] = Field(
+        default=None,
+        description=
+        "If set, at most moe_max_num_tokens tokens will be sent to torch.ops.trtllm.fused_moe at the same time. If the number of tokens exceeds moe_max_num_tokens, the input tensors will be split into chunks and a for loop will be used."
+    )
+
+    moe_load_balancer: Optional[Union[object, str]] = Field(
+        default=None,
+        description="Configuration for MoE load balancing.",
+        json_schema_extra={"type": "Union[MoeLoadBalancerConfig, str]"})
+
+    attn_backend: str = Field(default='TRTLLM',
+                              description="Attention backend to use.")
+
+    moe_backend: str = Field(default='CUTLASS',
+                             description="MoE backend to use.")
+
+    mixed_sampler: bool = Field(
+        default=False,
+        description=
+        "If true, will iterate over sampling_params of each request and use the corresponding sampling strategy, e.g. top-k, top-p, etc."
+    )
+
+    enable_trtllm_sampler: bool = Field(
+        default=False,
+        description=
+        "If true, will use the TRTLLM sampler instead of the PyTorch sampler. The TRTLLM sampler has a wide coverage of sampling strategies."
+    )
+
+    kv_cache_dtype: str = Field(default="auto",
+                                description="Data type for KV cache.")
+
+    use_kv_cache: bool = Field(default=True,
+                               description="Whether to use KV cache.")
+
+    enable_iter_perf_stats: bool = Field(
+        default=False, description="Enable iteration performance statistics.")
+
+    enable_iter_req_stats: bool = Field(
+        default=False,
+        description=
+        "If true, enables per request stats per iteration. Must also set enable_iter_perf_stats to true to get request stats."
+    )
+
+    print_iter_log: bool = Field(default=False,
+                                 description="Print iteration logs.")
+
+    torch_compile_enabled: bool = Field(
+        default=False, description="Enable torch.compile optimization.")
+
+    torch_compile_fullgraph: bool = Field(
+        default=True,
+        description="Enable full graph compilation in torch.compile.")
+
+    torch_compile_inductor_enabled: bool = Field(
+        default=False, description="Enable inductor backend in torch.compile.")
+
+    torch_compile_piecewise_cuda_graph: bool = Field(
+        default=False,
+        description="Enable piecewise CUDA graph in torch.compile.")
+
+    torch_compile_enable_userbuffers: bool = Field(
+        default=True,
+        description=
+        "When torch compile is enabled, userbuffers is enabled by default.")
+
+    autotuner_enabled: bool = Field(
+        default=True,
+        description="Enable autotuner only when torch compile is enabled.")
+
+    enable_layerwise_nvtx_marker: bool = Field(
+        default=False, description="If true, enable layerwise nvtx marker.")
+
+    auto_deploy_config: Optional[object] = Field(
+        default=None,
+        description="Auto deploy config.",
+        exclude_from_json=True,
+        json_schema_extra={"type": f"Optional[AutoDeployConfig]"})
+
+    load_format: Union[str, LoadFormat] = Field(
+        default=LoadFormat.AUTO,
+        description=
+        "How to load the model weights. By default, detect the weight type from the model checkpoint."
+    )
+
+    enable_min_latency: bool = Field(
+        default=False,
+        description=
+        "If true, enable min-latency mode. Currently only used for Llama4.",
+    )
+
+
    
+[docs]
+    @field_validator('load_format', mode='before')
+    @classmethod
+    def convert_load_format(cls, v):
+        if isinstance(v, LoadFormat):
+            return v
+        load_format = v.upper()
+        if load_format not in LoadFormat.__members__:
+            raise ValueError(f"Invalid LoadFormat: {v}")
+        return LoadFormat[load_format]
    
+
+
+    # Extra resource managers to use in addition to the KV cache manager.
+    # Each manager's prepare_resources method is called before the forward pass,
+    # and update_resources() is called after the pass finishes. free_resources()
+    # is called when a request finishes. The KV cache manager is guaranteed to
+    # be invoked after all of these extra managers in all stages.
+    _extra_resource_managers: Dict[str,
+                                   object] = PrivateAttr(default_factory=dict, )
+
+    @property
+    def extra_resource_managers(self) -> Dict[str, object]:
+        return self._extra_resource_managers
+
+    @extra_resource_managers.setter
+    def extra_resource_managers(self, value: Dict[str, object]) -> None:
+        self._extra_resource_managers = value
+
+
    
+[docs]
+    @print_traceback_on_error
+    def model_post_init(self, __context):
+        from .._torch.model_config import MoeLoadBalancerConfig
+
+        super().model_post_init(__context)
+        self.model_format = _ModelFormatKind.HF
+
+        if isinstance(self.moe_load_balancer, str):
+            if not os.path.exists(self.moe_load_balancer):
+                raise FileNotFoundError(
+                    f"MoE load balancer config file not found: {self.moe_load_balancer}"
+                )
+            try:
+                with open(self.moe_load_balancer) as f:
+                    moe_load_balancer_config = yaml.safe_load(f)
+                self.moe_load_balancer = MoeLoadBalancerConfig(
+                    **moe_load_balancer_config)
+            except Exception as e:
+                raise ValueError(
+                    f"Failed to load MoE load balancer config file: {self.moe_load_balancer}"
+                ) from e
    
+
+
+    # TODO: Remove this after the PyTorch backend is fully migrated to TorchLlmArgs from ExecutorConfig
+
    
+[docs]
+    def get_pytorch_backend_config(self) -> "PyTorchConfig":
+        from tensorrt_llm._torch.pyexecutor.config import PyTorchConfig
+
+        # TODO: Remove this after the PyTorch backend is fully migrated to TorchLlmArgs from ExecutorConfig
+        # Just a WAR to support the auto_deploy
+        if self.auto_deploy_config is not None:
+            return self.auto_deploy_config
+
+        return PyTorchConfig(
+            extra_resource_managers=self.extra_resource_managers,
+            use_cuda_graph=self.use_cuda_graph,
+            cuda_graph_batch_sizes=self.cuda_graph_batch_sizes,
+            cuda_graph_max_batch_size=self.cuda_graph_max_batch_size,
+            cuda_graph_padding_enabled=self.cuda_graph_padding_enabled,
+            disable_overlap_scheduler=self.disable_overlap_scheduler,
+            moe_max_num_tokens=self.moe_max_num_tokens,
+            moe_load_balancer=self.moe_load_balancer,
+            attn_backend=self.attn_backend,
+            moe_backend=self.moe_backend,
+            mixed_sampler=self.mixed_sampler,
+            enable_trtllm_sampler=self.enable_trtllm_sampler,
+            kv_cache_dtype=self.kv_cache_dtype,
+            use_kv_cache=self.use_kv_cache,
+            enable_iter_perf_stats=self.enable_iter_perf_stats,
+            enable_iter_req_stats=self.enable_iter_req_stats,
+            print_iter_log=self.print_iter_log,
+            torch_compile_enabled=self.torch_compile_enabled,
+            torch_compile_fullgraph=self.torch_compile_fullgraph,
+            torch_compile_inductor_enabled=self.torch_compile_inductor_enabled,
+            torch_compile_piecewise_cuda_graph=self.
+            torch_compile_piecewise_cuda_graph,
+            torch_compile_enable_userbuffers=self.
+            torch_compile_enable_userbuffers,
+            autotuner_enabled=self.autotuner_enabled,
+            enable_layerwise_nvtx_marker=self.enable_layerwise_nvtx_marker,
+            load_format=self.load_format,
+            enable_min_latency=self.enable_min_latency)
    
+
+
+
    
+[docs]
+    @field_validator('cuda_graph_max_batch_size')
+    @classmethod
+    def validate_cuda_graph_max_batch_size(cls, v):
+        """Validate cuda_graph_max_batch_size is non-negative."""
+        if v < 0:
+            raise ValueError("cuda_graph_max_batch_size must be non-negative")
+        return v
    
+
+
+    @staticmethod
+    def _generate_cuda_graph_batch_sizes(max_batch_size: int,
+                                         padding_enabled: bool) -> List[int]:
+        """Generate a list of batch sizes for CUDA graphs.
+
+        Args:
+            max_batch_size: Maximum batch size to generate up to
+            padding_enabled: Whether padding is enabled, which affects the batch size distribution
+
+        Returns:
+            List of batch sizes to create CUDA graphs for
+        """
+        if padding_enabled:
+            batch_sizes = [1, 2, 4] + [i * 8 for i in range(1, 17)]
+        else:
+            batch_sizes = list(range(1, 32)) + [32, 64, 128]
+
+        # Add powers of 2 up to max_batch_size
+        batch_sizes += [
+            2**i for i in range(8, math.floor(math.log(max_batch_size, 2)))
+        ]
+
+        # Filter and sort batch sizes
+        batch_sizes = sorted(
+            [size for size in batch_sizes if size <= max_batch_size])
+
+        # Add max_batch_size if not already included
+        if max_batch_size != batch_sizes[-1]:
+            batch_sizes.append(max_batch_size)
+
+        return batch_sizes
+
+
    
+[docs]
+    @model_validator(mode='after')
+    def validate_cuda_graph_config(self) -> 'TorchLlmArgs':
+        """Validate CUDA graph configuration.
+
+        Ensures that:
+        1. If cuda_graph_batch_sizes is provided, cuda_graph_max_batch_size must be 0
+        2. If cuda_graph_batch_sizes is not provided, it is generated based on cuda_graph_max_batch_size
+        3. If both are provided, cuda_graph_batch_sizes must match the generated values
+        """
+        if self.cuda_graph_batch_sizes is not None:
+            self.cuda_graph_batch_sizes = sorted(self.cuda_graph_batch_sizes)
+            if self.cuda_graph_max_batch_size != 0:
+                if self.cuda_graph_batch_sizes != self._generate_cuda_graph_batch_sizes(
+                        self.cuda_graph_max_batch_size,
+                        self.cuda_graph_padding_enabled):
+                    raise ValueError(
+                        "Please don't set both cuda_graph_batch_sizes "
+                        "and cuda_graph_max_batch_size.\n"
+                        f"cuda_graph_batch_sizes: {self.cuda_graph_batch_sizes}, "
+                        f"cuda_graph_max_batch_size: {self.cuda_graph_max_batch_size}"
+                    )
+            else:
+                self.cuda_graph_max_batch_size = max(
+                    self.cuda_graph_batch_sizes)
+        else:
+            max_batch_size = self.cuda_graph_max_batch_size or 128
+            generated_sizes = self._generate_cuda_graph_batch_sizes(
+                max_batch_size, self.cuda_graph_padding_enabled)
+            self.cuda_graph_batch_sizes = generated_sizes
+            self.cuda_graph_max_batch_size = max_batch_size
+
+        return self
    
+
    
+
+
+
 def update_llm_args_with_extra_dict(
         llm_args: Dict,
         llm_args_dict: Dict,
@@ -2126,6 +2527,15 @@
   
    
 
    
       
+        
    
+
    
+  
+  
    Last updated on June 03, 2025.
    
+  
+  
    This page is generated by TensorRT-LLM commit 9ae2ce6.
    
+  
+
    
+      
     
    
   
   
diff --git a/latest/_modules/tensorrt_llm/llmapi/mpi_session.html b/latest/_modules/tensorrt_llm/llmapi/mpi_session.html
index 158997e540..22c585bdba 100644
--- a/latest/_modules/tensorrt_llm/llmapi/mpi_session.html
+++ b/latest/_modules/tensorrt_llm/llmapi/mpi_session.html
@@ -50,7 +50,7 @@
     
@@ -60,7 +60,7 @@
 
   
   
-  
+  
 
 
   
@@ -329,6 +329,7 @@
 
  • Generate Text Asynchronously
    • 
 
  • Distributed LLM Generation
    • 
 
  • Control generated text using logits processor
    • 
+
  • Generate Text Using Eagle2 Decoding
    • 
 
  • Get KV Cache Events
    • 
 
  • Generate Text Using Lookahead Decoding
    • 
 
  • Generation with Quantization
    • 
@@ -350,6 +351,7 @@
 
  • Generate Text Asynchronously
    • 
 
  • Distributed LLM Generation
    • 
 
  • Control generated text using logits processor
    • 
+
  • Generate Text Using Eagle2 Decoding
    • 
 
  • Get KV Cache Events
    • 
 
  • Generate Text Using Lookahead Decoding
    • 
 
  • Generation with Quantization
    • 
@@ -414,6 +416,7 @@
 
  • Graph Rewriting Module
    • 
 
  • Run gpt-2b + LoRA using Executor / cpp runtime
    • 
 
  • Expert Parallelism in TensorRT-LLM
    • 
+
  • KV Cache Management: Pools, Blocks, and Events
    • 
 
  • KV cache reuse
    • 
 
  • Speculative Sampling
    • 
 
  • Disaggregated-Service (experimental)
    • 
@@ -448,6 +451,7 @@
 
  • Speed up inference with SOTA quantization techniques in TRT-LLM
    • 
 
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
    • 
 
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • 
+
  • DeepSeek R1 MTP Implementation and Optimization
    •
    @@ -688,9 +692,6 @@
    [docs] def __init__(self, comm=None, n_workers: int = 1): - if not external_mpi_comm_available(n_workers): - raise RuntimeError('The LLM instance should be launched by mpirun.') - self.comm = comm self.n_workers = n_workers self.thread_pool: Optional[ThreadPoolExecutor] = None @@ -1147,6 +1148,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/baichuan/model.html b/latest/_modules/tensorrt_llm/models/baichuan/model.html index 121f134443..d90b92f478 100644 --- a/latest/_modules/tensorrt_llm/models/baichuan/model.html +++ b/latest/_modules/tensorrt_llm/models/baichuan/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -873,6 +877,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/bert/model.html b/latest/_modules/tensorrt_llm/models/bert/model.html index 8e7fe4390a..c02ec73408 100644 --- a/latest/_modules/tensorrt_llm/models/bert/model.html +++ b/latest/_modules/tensorrt_llm/models/bert/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1177,6 +1181,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/bloom/model.html b/latest/_modules/tensorrt_llm/models/bloom/model.html index dcc93556a6..3f1e04df38 100644 --- a/latest/_modules/tensorrt_llm/models/bloom/model.html +++ b/latest/_modules/tensorrt_llm/models/bloom/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -785,6 +789,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/chatglm/config.html b/latest/_modules/tensorrt_llm/models/chatglm/config.html index 4e84f1d005..82d441fa93 100644 --- a/latest/_modules/tensorrt_llm/models/chatglm/config.html +++ b/latest/_modules/tensorrt_llm/models/chatglm/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -802,6 +806,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/chatglm/model.html b/latest/_modules/tensorrt_llm/models/chatglm/model.html index bb7c9e4800..079b8e1f18 100644 --- a/latest/_modules/tensorrt_llm/models/chatglm/model.html +++ b/latest/_modules/tensorrt_llm/models/chatglm/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1001,6 +1005,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/clip/model.html b/latest/_modules/tensorrt_llm/models/clip/model.html index da2bb70e0c..1a6b4656ce 100644 --- a/latest/_modules/tensorrt_llm/models/clip/model.html +++ b/latest/_modules/tensorrt_llm/models/clip/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -830,6 +834,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/cogvlm/config.html b/latest/_modules/tensorrt_llm/models/cogvlm/config.html index d2935fa0bf..725efe1916 100644 --- a/latest/_modules/tensorrt_llm/models/cogvlm/config.html +++ b/latest/_modules/tensorrt_llm/models/cogvlm/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -661,6 +665,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/cogvlm/model.html b/latest/_modules/tensorrt_llm/models/cogvlm/model.html index 5cf1d41816..5a72b28a35 100644 --- a/latest/_modules/tensorrt_llm/models/cogvlm/model.html +++ b/latest/_modules/tensorrt_llm/models/cogvlm/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -914,6 +918,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/commandr/model.html b/latest/_modules/tensorrt_llm/models/commandr/model.html index 2fa81a7940..344cbc789e 100644 --- a/latest/_modules/tensorrt_llm/models/commandr/model.html +++ b/latest/_modules/tensorrt_llm/models/commandr/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -812,6 +816,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/dbrx/config.html b/latest/_modules/tensorrt_llm/models/dbrx/config.html index 7927331bb3..914d526e83 100644 --- a/latest/_modules/tensorrt_llm/models/dbrx/config.html +++ b/latest/_modules/tensorrt_llm/models/dbrx/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -676,6 +680,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/dbrx/model.html b/latest/_modules/tensorrt_llm/models/dbrx/model.html index cba710ae0d..abecf36478 100644 --- a/latest/_modules/tensorrt_llm/models/dbrx/model.html +++ b/latest/_modules/tensorrt_llm/models/dbrx/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -802,6 +806,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/deepseek_v1/model.html b/latest/_modules/tensorrt_llm/models/deepseek_v1/model.html index fc6495ebef..e1e4f5805c 100644 --- a/latest/_modules/tensorrt_llm/models/deepseek_v1/model.html +++ b/latest/_modules/tensorrt_llm/models/deepseek_v1/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -896,6 +900,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/deepseek_v2/model.html b/latest/_modules/tensorrt_llm/models/deepseek_v2/model.html index 6bb2bd45fa..237dedb2d0 100644 --- a/latest/_modules/tensorrt_llm/models/deepseek_v2/model.html +++ b/latest/_modules/tensorrt_llm/models/deepseek_v2/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -978,6 +982,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/dit/model.html b/latest/_modules/tensorrt_llm/models/dit/model.html index 4fe93289e6..2e62c27bf2 100644 --- a/latest/_modules/tensorrt_llm/models/dit/model.html +++ b/latest/_modules/tensorrt_llm/models/dit/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1014,6 +1018,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/eagle/model.html b/latest/_modules/tensorrt_llm/models/eagle/model.html index 4b0c34cf57..8059e80fca 100644 --- a/latest/_modules/tensorrt_llm/models/eagle/model.html +++ b/latest/_modules/tensorrt_llm/models/eagle/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1950,6 +1954,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/enc_dec/model.html b/latest/_modules/tensorrt_llm/models/enc_dec/model.html index a9ded9df0d..fac8df8da0 100644 --- a/latest/_modules/tensorrt_llm/models/enc_dec/model.html +++ b/latest/_modules/tensorrt_llm/models/enc_dec/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -2855,6 +2859,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/falcon/config.html b/latest/_modules/tensorrt_llm/models/falcon/config.html index 6e615485c3..11d7eeb7ef 100644 --- a/latest/_modules/tensorrt_llm/models/falcon/config.html +++ b/latest/_modules/tensorrt_llm/models/falcon/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -737,6 +741,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/falcon/model.html b/latest/_modules/tensorrt_llm/models/falcon/model.html index 812fc1ee3e..85b1be8036 100644 --- a/latest/_modules/tensorrt_llm/models/falcon/model.html +++ b/latest/_modules/tensorrt_llm/models/falcon/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -899,6 +903,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gemma/config.html b/latest/_modules/tensorrt_llm/models/gemma/config.html index e92d1f837c..961dd54eb6 100644 --- a/latest/_modules/tensorrt_llm/models/gemma/config.html +++ b/latest/_modules/tensorrt_llm/models/gemma/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -827,6 +831,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gemma/model.html b/latest/_modules/tensorrt_llm/models/gemma/model.html index 09de913de8..e80fa36866 100644 --- a/latest/_modules/tensorrt_llm/models/gemma/model.html +++ b/latest/_modules/tensorrt_llm/models/gemma/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1019,6 +1023,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gpt/config.html b/latest/_modules/tensorrt_llm/models/gpt/config.html index c4d6aaa95f..20156f06cd 100644 --- a/latest/_modules/tensorrt_llm/models/gpt/config.html +++ b/latest/_modules/tensorrt_llm/models/gpt/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -946,6 +950,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gpt/model.html b/latest/_modules/tensorrt_llm/models/gpt/model.html index db39c0706f..db4d419722 100644 --- a/latest/_modules/tensorrt_llm/models/gpt/model.html +++ b/latest/_modules/tensorrt_llm/models/gpt/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1055,6 +1059,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gptj/config.html b/latest/_modules/tensorrt_llm/models/gptj/config.html index 0e030d97e4..fa05554092 100644 --- a/latest/_modules/tensorrt_llm/models/gptj/config.html +++ b/latest/_modules/tensorrt_llm/models/gptj/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -675,6 +679,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gptj/model.html b/latest/_modules/tensorrt_llm/models/gptj/model.html index fc4a558791..7082784cc5 100644 --- a/latest/_modules/tensorrt_llm/models/gptj/model.html +++ b/latest/_modules/tensorrt_llm/models/gptj/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -827,6 +831,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/gptneox/model.html b/latest/_modules/tensorrt_llm/models/gptneox/model.html index 1f0eac083d..03c4ff698e 100644 --- a/latest/_modules/tensorrt_llm/models/gptneox/model.html +++ b/latest/_modules/tensorrt_llm/models/gptneox/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -767,6 +771,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/llama/config.html b/latest/_modules/tensorrt_llm/models/llama/config.html index a1ffd20cef..8ec04ebf96 100644 --- a/latest/_modules/tensorrt_llm/models/llama/config.html +++ b/latest/_modules/tensorrt_llm/models/llama/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -901,6 +905,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/llama/model.html b/latest/_modules/tensorrt_llm/models/llama/model.html index f4cb58b7a7..e0986b01f8 100644 --- a/latest/_modules/tensorrt_llm/models/llama/model.html +++ b/latest/_modules/tensorrt_llm/models/llama/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1249,6 +1253,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/mamba/model.html b/latest/_modules/tensorrt_llm/models/mamba/model.html index 9cbde2eb0e..0440f62009 100644 --- a/latest/_modules/tensorrt_llm/models/mamba/model.html +++ b/latest/_modules/tensorrt_llm/models/mamba/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1094,6 +1098,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/medusa/config.html b/latest/_modules/tensorrt_llm/models/medusa/config.html index 32ef36311e..8c12f3cecb 100644 --- a/latest/_modules/tensorrt_llm/models/medusa/config.html +++ b/latest/_modules/tensorrt_llm/models/medusa/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -734,6 +738,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/medusa/model.html b/latest/_modules/tensorrt_llm/models/medusa/model.html index d488c8a280..bc80024bbc 100644 --- a/latest/_modules/tensorrt_llm/models/medusa/model.html +++ b/latest/_modules/tensorrt_llm/models/medusa/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -884,6 +888,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/mllama/model.html b/latest/_modules/tensorrt_llm/models/mllama/model.html index ec693bdaf4..5cbf6a0e7f 100644 --- a/latest/_modules/tensorrt_llm/models/mllama/model.html +++ b/latest/_modules/tensorrt_llm/models/mllama/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -2195,6 +2199,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/mmdit_sd3/model.html b/latest/_modules/tensorrt_llm/models/mmdit_sd3/model.html index f9826929aa..ce3902bdff 100644 --- a/latest/_modules/tensorrt_llm/models/mmdit_sd3/model.html +++ b/latest/_modules/tensorrt_llm/models/mmdit_sd3/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1261,6 +1265,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/modeling_utils.html b/latest/_modules/tensorrt_llm/models/modeling_utils.html index edab0f27f1..3ca5f1f7ea 100644 --- a/latest/_modules/tensorrt_llm/models/modeling_utils.html +++ b/latest/_modules/tensorrt_llm/models/modeling_utils.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -551,7 +555,7 @@ WeightOnlyQuantRowLinear) from ..quantization.mode import (KV_CACHE_QUANT_ALGO_LIST, QUANT_ALGO_LIST, W8A8_SQ_PLUGIN_LIST, QuantAlgo) -from ..quantization.utils.fp4_utils import float4_sf_dtype +from ..quantization.utils import fp4_utils from ..top_model_mixin import TopModelMixin from .convert_utils import weight_only_quantize_dict from .generation_mixin import GenerationMixin @@ -603,6 +607,7 @@ LOOKAHEAD_DECODING = auto() EXPLICIT_DRAFT_TOKENS = auto() EAGLE = auto() + NGRAM = auto()
    [docs] @@ -620,6 +625,8 @@ return SpeculativeDecodingMode.EXPLICIT_DRAFT_TOKENS elif args.speculative_decoding_mode == "eagle": return SpeculativeDecodingMode.EAGLE + elif args.speculative_decoding_mode == "ngram": + return SpeculativeDecodingMode.NGRAM else: assert False, "Unknown speculative_decoding_mode " + args.speculative_decoding_mode
    @@ -2389,15 +2396,18 @@ # Interleave block scale for NVFP4 plugin. for name in list(weights): if name.endswith('weights_scaling_factor'): - ori_shape = weights[name].shape + out_features, in_features = weights[name].shape + nrows = fp4_utils.pad_up(out_features, 128) + ncols = fp4_utils.pad_up(in_features, 4) new_name = name.replace('weights_scaling_factor', 'weights_block_scaling_factor') weights[new_name] = weights[name] weights[ new_name + "_interleaved"] = torch.ops.tensorrt_llm.nvfp4_block_scale_interleave( - weights[name].view(float4_sf_dtype).cpu().contiguous( - )).reshape(ori_shape).view(float4_sf_dtype) + weights[name].view(fp4_utils.float4_sf_dtype).cpu( + ).contiguous()).reshape(nrows, ncols).view( + fp4_utils.float4_sf_dtype) weights.pop(name) if name.endswith('weights_scaling_factor_2'): new_name = name.replace('weights_scaling_factor_2', @@ -2650,6 +2660,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/mpt/model.html b/latest/_modules/tensorrt_llm/models/mpt/model.html index fe0ee997ad..3f9a382258 100644 --- a/latest/_modules/tensorrt_llm/models/mpt/model.html +++ b/latest/_modules/tensorrt_llm/models/mpt/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -799,6 +803,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/multimodal_encoders/config.html b/latest/_modules/tensorrt_llm/models/multimodal_encoders/config.html index 590c33e21c..7bd34b393a 100644 --- a/latest/_modules/tensorrt_llm/models/multimodal_encoders/config.html +++ b/latest/_modules/tensorrt_llm/models/multimodal_encoders/config.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -733,6 +737,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/multimodal_encoders/model.html b/latest/_modules/tensorrt_llm/models/multimodal_encoders/model.html index da62711161..5fb65b6437 100644 --- a/latest/_modules/tensorrt_llm/models/multimodal_encoders/model.html +++ b/latest/_modules/tensorrt_llm/models/multimodal_encoders/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -801,6 +805,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/opt/model.html b/latest/_modules/tensorrt_llm/models/opt/model.html index 18c335588e..64822e5f83 100644 --- a/latest/_modules/tensorrt_llm/models/opt/model.html +++ b/latest/_modules/tensorrt_llm/models/opt/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -804,6 +808,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/phi/model.html b/latest/_modules/tensorrt_llm/models/phi/model.html index 542546f2cb..b5d512d010 100644 --- a/latest/_modules/tensorrt_llm/models/phi/model.html +++ b/latest/_modules/tensorrt_llm/models/phi/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -848,6 +852,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/phi3/model.html b/latest/_modules/tensorrt_llm/models/phi3/model.html index 2f45ed23d2..d9f3c283a4 100644 --- a/latest/_modules/tensorrt_llm/models/phi3/model.html +++ b/latest/_modules/tensorrt_llm/models/phi3/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -944,6 +948,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/recurrentgemma/model.html b/latest/_modules/tensorrt_llm/models/recurrentgemma/model.html index 6ad6b4c9d7..1e6a02647c 100644 --- a/latest/_modules/tensorrt_llm/models/recurrentgemma/model.html +++ b/latest/_modules/tensorrt_llm/models/recurrentgemma/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1247,6 +1251,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/models/redrafter/model.html b/latest/_modules/tensorrt_llm/models/redrafter/model.html index 82d3430574..ff2ecd37dd 100644 --- a/latest/_modules/tensorrt_llm/models/redrafter/model.html +++ b/latest/_modules/tensorrt_llm/models/redrafter/model.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -919,6 +923,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/plugin/plugin.html b/latest/_modules/tensorrt_llm/plugin/plugin.html index 61b6be5a89..ab4d18e833 100644 --- a/latest/_modules/tensorrt_llm/plugin/plugin.html +++ b/latest/_modules/tensorrt_llm/plugin/plugin.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -533,7 +537,8 @@ from .._ipc_utils import IpcMemory, can_access_peer from .._utils import get_sm_version from ..bindings.internal.runtime import (lamport_initialize, - lamport_initialize_all) + lamport_initialize_all, + max_workspace_size_lowprecision) from ..logger import logger from ..mapping import Mapping @@ -1191,7 +1196,7 @@ Then, each instance of allreduce will reference that tensor automatically. """ POINTERS_PER_RANK = 7 - POINTERS_OF_COUNTER = 2 + POINTERS_OF_COUNTER = 3 def __init__(self) -> None: self.workspace: Optional[Tensor] = None @@ -1225,6 +1230,17 @@ return 16_000_000 return 8_000_000 + @staticmethod + def max_workspace_size_lowprecision(tp_size: int) -> int: + return max_workspace_size_lowprecision(tp_size) + + @staticmethod + def initialize_lowprecision_buffers(workspace: "torch.tensor", + tp_size: int) -> None: + import torch + return torch.ops.trtllm.initialize_static_lowprecision_buffers( + workspace, tp_size) + @staticmethod def allocate_workspace(mapping: Mapping, size: int) -> Tuple[List[IpcMemory], "torch.tensor"]: @@ -1239,11 +1255,11 @@ ipc_buffers_pong = IpcMemory(mapping, ipc_buffers_size, is_p2p_supported) ipc_barriers_in = IpcMemory( - mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2, - is_p2p_supported) + mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2 * + mapping.tp_size, is_p2p_supported) ipc_barriers_out = IpcMemory( - mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2, - is_p2p_supported) + mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2 * + mapping.tp_size, is_p2p_supported) lamport_buffers_size = 1 if force_deterministic else size * mapping.tp_size lamport_buffers_0 = IpcMemory(mapping, lamport_buffers_size, is_p2p_supported) @@ -1261,16 +1277,55 @@ lamport_buffers_size, ) buffers = [ - ipc_buffers_ping, ipc_buffers_pong, ipc_barriers_in, - ipc_barriers_out, lamport_buffers_0, lamport_buffers_1, - lamport_buffers_2 + ipc_buffers_ping, + ipc_buffers_pong, + ipc_barriers_in, + ipc_barriers_out, + lamport_buffers_0, + lamport_buffers_1, + lamport_buffers_2, + # Start from 1 since 0 represents released state for barrier at the beginning of the all_reduce. + # The last element is the barrier flag counter. + torch.tensor([1, 1, 0], dtype=torch.int64, device="cuda") ] return buffers, torch.tensor( ipc_buffers_ping.serialize() + ipc_buffers_pong.serialize() + ipc_barriers_in.serialize() + ipc_barriers_out.serialize() + lamport_buffers_0.serialize() + lamport_buffers_1.serialize() + - lamport_buffers_2.serialize() + [0] + [0], + lamport_buffers_2.serialize() + [buffers[-1].data_ptr()] + + [buffers[-1][1:].data_ptr()] + [buffers[-1][2:].data_ptr()], + dtype=torch.int64, + device="cpu") + + @staticmethod + def allocate_lowprecision_workspace( + mapping: Mapping, + size: int) -> Tuple[List[IpcMemory], "torch.tensor"]: + import torch + + # Force pull mode and disable lamport when force deterministic is enabled, for reducing device memory usage. + is_p2p_supported = can_access_peer(mapping) + ipc_buffers_size = size + ipc_buffers_ping = IpcMemory(mapping, ipc_buffers_size, + is_p2p_supported) + ipc_buffers_pong = IpcMemory(mapping, ipc_buffers_size, + is_p2p_supported) + ipc_barriers_in = IpcMemory( + mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2, + is_p2p_supported) + ipc_barriers_out = IpcMemory( + mapping, IpcMemory.IPC_BARRIERS_SIZE_PER_GPU * mapping.tp_size * 2, + is_p2p_supported) + buffers = [ + ipc_buffers_ping, ipc_buffers_pong, ipc_barriers_in, + ipc_barriers_out + ] + + return buffers, torch.tensor( + ipc_buffers_ping.serialize() + ipc_buffers_pong.serialize() + + ipc_barriers_in.serialize() + ipc_barriers_out.serialize() + [0] + + [0], dtype=torch.int64, device="cpu") @@ -1424,6 +1479,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/quantization/mode.html b/latest/_modules/tensorrt_llm/quantization/mode.html index 18d684332b..1a57fca377 100644 --- a/latest/_modules/tensorrt_llm/quantization/mode.html +++ b/latest/_modules/tensorrt_llm/quantization/mode.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1027,6 +1031,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/quantization/quantize_by_modelopt.html b/latest/_modules/tensorrt_llm/quantization/quantize_by_modelopt.html index 7fbb7ddb31..bffc8235ee 100644 --- a/latest/_modules/tensorrt_llm/quantization/quantize_by_modelopt.html +++ b/latest/_modules/tensorrt_llm/quantization/quantize_by_modelopt.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1888,6 +1892,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/enc_dec_model_runner.html b/latest/_modules/tensorrt_llm/runtime/enc_dec_model_runner.html index 6543130e96..14a8374cb0 100644 --- a/latest/_modules/tensorrt_llm/runtime/enc_dec_model_runner.html +++ b/latest/_modules/tensorrt_llm/runtime/enc_dec_model_runner.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1158,6 +1162,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/generation.html b/latest/_modules/tensorrt_llm/runtime/generation.html index bb9167849c..f55c97392a 100644 --- a/latest/_modules/tensorrt_llm/runtime/generation.html +++ b/latest/_modules/tensorrt_llm/runtime/generation.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -5446,6 +5450,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/kv_cache_manager.html b/latest/_modules/tensorrt_llm/runtime/kv_cache_manager.html index 122753aa16..84659692e3 100644 --- a/latest/_modules/tensorrt_llm/runtime/kv_cache_manager.html +++ b/latest/_modules/tensorrt_llm/runtime/kv_cache_manager.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1105,6 +1109,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/model_runner.html b/latest/_modules/tensorrt_llm/runtime/model_runner.html index 331b1818aa..2bb6e85224 100644 --- a/latest/_modules/tensorrt_llm/runtime/model_runner.html +++ b/latest/_modules/tensorrt_llm/runtime/model_runner.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1617,6 +1621,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/model_runner_cpp.html b/latest/_modules/tensorrt_llm/runtime/model_runner_cpp.html index b5efa1a9d0..5f129f5cef 100644 --- a/latest/_modules/tensorrt_llm/runtime/model_runner_cpp.html +++ b/latest/_modules/tensorrt_llm/runtime/model_runner_cpp.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -1010,7 +1014,9 @@ @property def num_layers(self) -> int: return self.model_config.num_layers( - self.world_config.pipeline_parallelism) + self.world_config.pipeline_parallelism, + self.world_config.pipeline_parallel_rank, + ) @property def max_sequence_length(self) -> int: @@ -1819,6 +1825,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/multimodal_model_runner.html b/latest/_modules/tensorrt_llm/runtime/multimodal_model_runner.html index d95fa8d16f..3418ba3496 100644 --- a/latest/_modules/tensorrt_llm/runtime/multimodal_model_runner.html +++ b/latest/_modules/tensorrt_llm/runtime/multimodal_model_runner.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -3347,6 +3351,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/runtime/session.html b/latest/_modules/tensorrt_llm/runtime/session.html index 201d6ff0f9..4c54b2be61 100644 --- a/latest/_modules/tensorrt_llm/runtime/session.html +++ b/latest/_modules/tensorrt_llm/runtime/session.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -965,6 +969,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_modules/tensorrt_llm/sampling_params.html b/latest/_modules/tensorrt_llm/sampling_params.html index b057c84293..24f7438145 100644 --- a/latest/_modules/tensorrt_llm/sampling_params.html +++ b/latest/_modules/tensorrt_llm/sampling_params.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -513,6 +517,7 @@ from pydantic import BaseModel from tensorrt_llm.bindings import executor as tllme +from tensorrt_llm.executor.serialization import register_approved_ipc_class
    @@ -579,6 +584,14 @@ """ pass # noqa + def __init_subclass__(cls, **kwargs): + """ + This method is called when a class inherits from LogitsProcessor. + """ + # Register subclass as an approved class for deserialization across IPC boundaries. + super().__init_subclass__(**kwargs) + register_approved_ipc_class(cls) + class BatchedLogitsProcessor(ABC): """Base class for batched logits processor. @@ -790,20 +803,18 @@ For instance, while the greedy decoding with n > 1 is capable in the Executor class of C++ runtime, the LLM API disallows such combination. ''' - if self.best_of is not None: - if self.best_of > 1 and self.best_of < self.n: - raise ValueError( - f'In beam search, best_of ({self.best_of}) must be ' - f'greater than or equal to n ({self.n}).') + if self.best_of < self.n: + raise ValueError( + f"best_of ({self.best_of}) cannot be less than n ({self.n})") - if (self.best_of > 1 and self._greedy_decoding and - not os.environ.get('TLLM_ALLOW_N_GREEDY_DECODING', None)): - raise ValueError( - f'Greedy decoding in the LLM API does not allow multiple ' - f'returns. Please set to best_of=1, got best_of={self.best_of}. ' - f'Please set to best_of=1 or set an environment variable ' - f'TLLM_ALLOW_N_GREEDY_DECODING=1 to allow best_of > 1 ' - f'under the greedy decoding.') + if (self.best_of > 1 and self._greedy_decoding + and not os.environ.get('TLLM_ALLOW_N_GREEDY_DECODING', None)): + raise ValueError( + f'Greedy decoding in the LLM API does not allow multiple ' + f'returns. Please set to best_of=1, got best_of={self.best_of}. ' + f'Please set to best_of=1 or set an environment variable ' + f'TLLM_ALLOW_N_GREEDY_DECODING=1 to allow best_of > 1 ' + f'under the greedy decoding.') if self.truncate_prompt_tokens is not None and self.truncate_prompt_tokens < 1: raise ValueError( @@ -1092,6 +1103,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/_sources/_cpp_gen/executor.rst.txt b/latest/_sources/_cpp_gen/executor.rst.txt index 08d47843b4..d3ca9cd473 100644 --- a/latest/_sources/_cpp_gen/executor.rst.txt +++ b/latest/_sources/_cpp_gen/executor.rst.txt @@ -16,6 +16,12 @@ ________ .. doxygenfile:: tensor.h :project: TensorRT-LLM +transferAgent.h +_______________ + +.. doxygenfile:: transferAgent.h + :project: TensorRT-LLM + serialization.h _______________ diff --git a/latest/_sources/advanced/kv-cache-management.md.txt b/latest/_sources/advanced/kv-cache-management.md.txt new file mode 100644 index 0000000000..f4506d6ee9 --- /dev/null +++ b/latest/_sources/advanced/kv-cache-management.md.txt @@ -0,0 +1,75 @@ +(kv-cache-management)= + +# KV Cache Management: Pools, Blocks, and Events + +This document provides an overview of the internal hierarchy and event system for paged KV cache management, as implemented in the TensorRT-LLM codebase. + +For more information on KV cache reuse see [KV cache reuse](kv-cache-reuse.md). + +--- + +## Hierarchy: Pool, Block, and Page + +### **Block** +- **Definition:** The smallest unit of KV cache allocation. A `KVCacheBlock` holds metadata (not the actual data) for a chunk of KV cache. +- **Purpose:** Each block represents a fixed number of tokens' worth of KV data (can be specified by `tokens_per_block` parameter). +- **Usage:** Blocks are allocated, reused, or evicted as sequences are processed. + +### **Page** +- **Definition:** In this codebase, "page" is often used interchangeably with "block" (as in "paged KV cache"), but technically, a page could refer to a memory page (hardware-level), while a block is a logical unit for the cache. +- **In Practice:** The code uses "block" as the main unit; "page" is not a distinct class or struct. + +### **Pool** +- **Definition:** A pool is a contiguous memory buffer (or set of buffers) that holds the actual KV data for one or more layers. +- **Types:** There are primary pools (fast GPU memory) and secondary pools (slower, e.g., CPU or offload memory). +- **Organization:** Each pool can serve multiple layers that share the same KV head configuration. Pools are managed by `KVCacheBlockPool` and tracked in vectors in `WindowBlockManager`. +- **Block ↔ Pool:** Each block is an index into a pool; the pool provides the actual storage, while the block is the metadata handle. + +### **WindowBlockManager/BlockManager** + +TRT-LLM supports 2 complex features related to KV cache management: +1. **Variable Group-Query Attention (VGQA)** - i.e. a different `num_kv_heads` value for different layers. +2. **Variable Sliding Window Attention (VSWA)** - i.e. a different `attention_window_size` value for different layers. + +In order to support both of these features, the pool management works as described below. + +But in the simple, *most common case*, for most models, where +1. [MHA/MQA/Non-variable GQA](gpt-attention.md#multi-head-multi-query-and-group-query-attention), i.e., same `num_kv_heads` value for all layers, +2. Global attention/[SWA](gpt-attention.md#sliding-window-attention-cyclic-rolling-buffer-kv-cache), i.e., same `attention_window_size` value for all layers, + +only a *single* pool will be created within the structure described below. + +#### KV Cache Pool Management + +- **WindowBlockManager:** Manages blocks and pools for a specific attention window size. Within a `WindowBlockManager`, there can be multiple pools - each corresponding a unique number of KV heads - i.e., to support VGQA. +- **BlockManager:** Manages all `WindowBlockManager` instances, one per unique window size. + +**Hierarchy Summary:** +- **Pool** (memory buffer for KV data) + - Contains many blocks. +- **Blocks** (metadata for a chunk of the pool, each block = `tokens_per_block` tokens) + - (Optionally, blocks can be swapped between primary/secondary pools.) +- **BlockManager/WindowBlockManager**: Manage pools and blocks, handle allocation, reuse, and eviction. + +--- + +## Events in `KVCacheEventManager` + +The `KVCacheEventManager` is responsible for tracking and reporting significant changes in the state of the KV cache. Events are used for logging, debugging, or possibly for external monitoring. + +### **Types of Events** +- **Created Event:** When pools or blocks are created/allocated. +- **Updated Event:** When a block's state changes (e.g., moved between primary/secondary, priority updated). +- **Removed Event:** When a block is removed from the cache (evicted or released). +- **Stored Event:** When blocks are stored for potential reuse (e.g., after a sequence finishes and its blocks are reusable). + +### **What Triggers an Event?** +- **Allocation/Deallocation:** Creating or freeing memory pools or blocks. +- **Eviction/Reuse:** When a block is evicted, reused, or its priority changes. +- **Block Movement:** When a block is moved between memory levels (primary ↔ secondary). +- **Block Storage:** When blocks are stored for future reuse (e.g., after a sequence completes). + +**In summary:** +An "event" is any significant change in the lifecycle or state of a KV cache block or pool, tracked for monitoring, debugging, or optimization purposes. + +--- diff --git a/latest/_sources/advanced/lowprecision-pcie-allreduce.md.txt b/latest/_sources/advanced/lowprecision-pcie-allreduce.md.txt new file mode 100644 index 0000000000..57ca754c4e --- /dev/null +++ b/latest/_sources/advanced/lowprecision-pcie-allreduce.md.txt @@ -0,0 +1,65 @@ +# Low-Precision-AllReduce + +```{note} +Note: +This feature is optimized for PCIe-based GPU topologies and may affect model accuracy. Please evaluate precision impact for your specific workload. +``` + + +TRT-LLM supports `low-precision-allreduce`, a communication optimization that accelerates AllReduce operations in PCIe-based GPU environments. This feature quantizes FP16/BF16 data to FP8 during network transmission, reducing communication volume and improving performance. + +## Algorithm + +The Low-Precision-AllReduce algorithm works by: +1. Quantizing input FP16/BF16 tensors to FP8 format before network transmission + + + **Quantization details**: We use a "per-warp" quantization approach where each CUDA warp (32 threads) processes a batch of data. In each warp, 31 threads quantize FP16/BF16 values to FP8 e4m3 format (16 bytes per thread), while the last thread transmits a scalar value. This results in each warp collectively quantizing 496 elements plus one scalar at a time. + +2. Transmitting the quantized data through the network +3. Dequantizing received data back to the original precision +4. Performing the reduction operation + +In 8-GPU scenarios, this approach shifts the communication bottleneck from cross-NUMA QPI to the PCIe switch, resulting in better overall performance. + +## Topology Requirements + +![8x L20/L40s Node Architecture](images/8x_l20_L40S_node_architecture.png) + +Low-Precision-AllReduce is specifically designed for the topology shown above, where: +- Each node contains 2 NUMA domains +- Each NUMA domain has 4 GPUs connected via PCIe switch +- GPUs within the same NUMA node communicate via the PCIe switch + +**Important:** This optimization will not accelerate performance in different topologies (e.g., where each GPU is in a separate NUMA domain). + +## Usage + +The Low-Precision-AllReduce algorithm can be enabled in two ways: + +1. **Direct specification** in your code: +``` +AllReduce allreduce(mapping=mapping, strategy=AllReduceStrategy.LOWPRECISION); +``` +2. **Environment variable control** with AUTO strategy: +``` +// In your code +AllReduce allreduce(mapping=mapping, strategy=AllReduceStrategy.AUTO); +// Set environment variable before running +export FORCE_LOW_PRECISION_ALL_REDUCE_STRATEGY=1 +``` + +## Performance and Accuracy Considerations + +Low-Precision-AllReduce reduces communication volume by using FP8 data format for transmission. This optimization: +- Improves performance for large message sizes in PCIe-based topologies +- May slightly reduce numerical precision +- Automatically falls back to other strategies when no performance benefit is expected (e.g., with NVLink or small messages) + +Users should evaluate the precision impact on their specific models and workloads. + +## Environment Variables + +- `FORCE_LOW_PRECISION_ALL_REDUCE_STRATEGY`: When set to `1`, forces the use of low-precision algorithm with AUTO strategy. If the algorithm determines it cannot provide performance benefits, it will automatically fall back to other strategies. + +**Note**: When compiling TensorRT-LLM without enabling the `ENABLE_FP8` option, setting Low Precision allreduce will not take effect. diff --git a/latest/_sources/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md.txt b/latest/_sources/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md.txt index 7f90c391c0..d510209b4a 100644 --- a/latest/_sources/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md.txt +++ b/latest/_sources/blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md.txt @@ -134,9 +134,8 @@ To do the benchmark, run the following command: YOUR_DATA_PATH= cat >./extra-llm-api-config.yml<./extra-llm-api-config.yml < cat >./extra-llm-api-config.yml<./extra-llm-api-config.yml< +tech_blog1_model_overview ### Precision Strategy We have explored a mixed precision recipe, which provides a better tradeoff between accuracy and performance. @@ -84,7 +84,7 @@ We have also explored and introduced mixed parallel strategy on 8xB200 GPUs. Spe ### Everything in One Diagram Now let's put everything into one diagram, which represents a MoE layer from a decoding iteration. -tech_blog1_model_details +tech_blog1_model_details The modules in the diagram are: @@ -136,7 +136,7 @@ The modules in the diagram are: | Optimize CUTLASS Flow: Sparse Experts as GEMMs | 249 | The code is not open-source yet due to the dependency with internal base environment and we are planning to make it decoupled from internal base environment thus to be able to open-source in the future.| | Introduce EP4TP2 for better workload balance | 253 | Use `--tp 8 --ep 4` when benchmarking | | Introduce moe_backend=TRTLLM, EP2TP4 for better balance | 299 | [PR #4280](https://github.com/NVIDIA/TensorRT-LLM/pull/4280) | -| Optimize Fuse_A_GEMM and Router_GEMM | 340 | WIP: [PR #4115](https://github.com/NVIDIA/TensorRT-LLM/pull/4115) | +| Optimize Fuse_A_GEMM and Router_GEMM | 340 | WIP | | Relax Acceptance | **368** | [deepseek_v3#multi-token-prediction-mtp](https://github.com/NVIDIA/TensorRT-LLM/tree/main/examples/models/core/deepseek_v3#multi-token-prediction-mtp) | ### System Level optimizations @@ -195,7 +195,7 @@ We have introduced multi-streams based optimizations to hide some kernels' overh #### Sparse Experts as GEMMs (only works when moe_backend=CUTLASS) -tech_blog1_sparse_exp_as_a_gemm +tech_blog1_sparse_exp_as_a_gemm The existing CUTLASS-based Sparse Experts flow (illustrated in the figure) dispatches input tokens to their designated experts, then applies indexed local reduction on each expert's outputs before a global allreduce. Both dispatching and indexed local reduction incur high overhead in low-latency scenarios. To address this, we propose treating "Sparse Experts as GEMMs" by sending all tokens to each activated expert and masking out unneeded outputs before local reduction. Because grouped GEMMs are memory-bound, the extra computations from redundant tokens have minimal impact, effectively eliminating the costly dispatch and reduction overhead. @@ -229,12 +229,12 @@ We focus on optimizing two kinds of dense GEMMs: Fuse_A_GEMM and RouterGEMM, bec ##### Fuse_A_GEMM We developed a custom Fuse_A_GEMM that prefetches the majority of its weights into shared memory (enabled by PDL and overlapped with oneshot-AllReduce), significantly enhancing performance. The kernel shows substantial improvements over default GEMM implementation when num_tokens < 16. -tech_blog1_fuse_a_gemm +tech_blog1_fuse_a_gemm ##### RouterGEMM -By leveraging our internal AI code generator, we automatically generate an optimized RouterGEMM kernel, which delivers substantial improvements over the default GEMM implementation when [num_tokens <=30](https://github.com/NVIDIA/TensorRT-LLM/pull/4115/files#diff-006ae982200a5ef2b27f4aedb526025e64406d3c2fadde329ea745793fac04edR303:~:text=and%20hidden_states.-,size,-(0)) +By leveraging our internal AI code generator, we automatically generate an optimized RouterGEMM kernel, which delivers substantial improvements over the default GEMM implementation when num_tokens <=30. -tech_blog1_router_gemm +tech_blog1_router_gemm #### Kernel fusion Kernel fusion is necessary for min-latency scenario to reduce extra global memory write/read cost, and we support following fusion patterns now diff --git a/latest/_sources/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.md.txt b/latest/_sources/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.md.txt new file mode 100644 index 0000000000..0014f1c7f2 --- /dev/null +++ b/latest/_sources/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.md.txt @@ -0,0 +1,252 @@ +# DeepSeek R1 MTP Implementation and Optimization +by NVIDIA TensorRT-LLM team +## Table of Contents +- [MTP for inference](#mtp-for-inference) + - [Background](#background) + - [MTP Vanilla](#mtp-vanilla) + - [MTP Eagle](#mtp-eagle) +- [MTP implementation in TensorRT-LLM](#mtp-implementation-in-tensorrt-llm) + - [Basic Implementation](#basic-implementation) + - [MTP Modules](#mtp-modules) + - [Attention for MTP](#attention-for-mtp) + - [How to run DeepSeek models with MTP](#how-to-run-deepseek-models-with-mtp) +- [MTP optimization - Relaxed Acceptance](#mtp-optimization---relaxed-acceptance) + - [Relaxed Acceptance](#relaxed-acceptance) + - [How to run the DeepSeek-R1 model with Relaxed Acceptance](#how-to-run-the-deepseek-r1-model-with-relaxed-acceptance) +- [Evaluation](#evaluation) + - [Achieving speedup with MTP speculative decoding](#achieving-speedup-with-mtp-speculative-decoding) + - [Accuracy studies for Relaxed Acceptance](#accuracy-studies-for-relaxed-acceptance) +- [Future Works](#future-works) + - [Tree-based speculative decoding support](#tree-based-speculative-decoding-support) + - [Eagle3 support](#eagle3-support) + - [Fix known issues](#fix-known-issues) +- [Acknowledgment](#acknowledgment) + + +TensorRT-LLM achieves world-record inference performance for DeepSeek-R1 on NVIDIA Blackwell GPUs, where Multi-Token Prediction (MTP) delivers a significant speedup. In our [previous blog post](https://github.com/NVIDIA/TensorRT-LLM/blob/main/docs/source/blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs.md), we discussed the key optimizations that enable the outstanding inference latency of the DeepSeek-R1 model. This article dives deeper into the implementation and optimization of MTP in TensorRT-LLM. + +## MTP for inference +Inspired by a previous [research work](https://arxiv.org/pdf/2404.19737), MTP is designed to help the DeepSeek-V3 training. It adds additional MTP modules at the end of the main model and uses them to predict additional tokens. In this way, MTP can extend the prediction scope to multiple future tokens at each position to achieve better model accuracy. During inference, those MTP modules can also be used for speculative decoding to improve the generation latency further. In this section, we will introduce the MTP speculative decoding algorithm for LLM inference. + +### Background +Speculative decoding is a popular technique for faster and cost-effective LLM inference. It’s based on the premise that generating multiple future tokens(especially for decode phase which is less compute bound) is more efficient than processing a single token. Speculative decoding techniques usually divide the process into a low-cost draft stage and a parallelized verification stage. The draft stage predicts draft tokens by using a small model or a subset of layers in the main model. And the verification stage uses the main model to determine how many of these draft tokens to accept, which is far more efficient than generating one token per iteration. + +
    +
    + tech_blog2_verify_and_accept +
    +
    +

    Figure 1. Verification example

    + +Figure 1 shows an example of how to verify and accept those draft tokens. Assuming there are a total of 5 draft tokens “ABCDE”, we will extend them to the input token “G”, and input a total of 6 tokens to the main model. After sampling, we can get six different expected tokens, then compare the expected tokens with the draft tokens and accept the longest prefix matched tokens. In this example, the tokens “ABC” are matched. Because “H” is predicted by the main model and the corresponding input token “C” is already accepted, “H” will also be accepted. In this way, we can accept four tokens in a single iteration. MTP also uses this method to verify and accept draft tokens. +For the draft stage in MTP, there are two different MTP methods, MTP vanilla and MTP eagle. They can be used for different inference cases. + +### MTP Vanilla + +
    +
    + tech_blog2_mtp_vanilla +
    +
    +

    Figure 2. MTP Vanilla, where ti is the input token, di is the predicted draft token, K is the number of MTP modules, and hin is the hidden state of the n-th MTP module. Note that h0 means the hidden states of the main model. (Disclaimer: the figures adapted from the original DeepSeek V3 tech report)

    + + +MTP Vanilla method is more similar to the MTP training, and it sequentially uses different MTP modules to predict multiple draft tokens. This method can support model checkpoints with weights of multiple different MTP modules. And each MTP module will have its own KV cache. + +Figure 2 illustrates the MTP vanilla inference. In the context phase, assuming there are a total of four input tokens, we will get the output token $t_5$ and the hidden states after the main model forward. The output token will be appended to the input tokens, then we shift out the first token to get tokens from $t_2$ to $t_5$ as the input tokens of the first MTP module. The hidden states from the main model will be directly used as the input of the first MTP module to predict the first draft token. For the next several MTP modules, we will use the same method to prepare the inputs to predict the sequential draft tokens. + +In the generation phase, there will be a little difference. The predicted token $t_5$ and the draft tokens will be used as inputs for the main model. After the main model forward, we will do the verification to get the accepted tokens. In this example, assuming $j$ draft tokens $d_6$~$d_{j+5}$ are accepted. Then prepare the MTP module inputs. Different from the context phase, we will prepare input IDs and hidden states of a total of $K$ tokens before the last accepted token. In this example, the last accepted token is $t_{j+6}$. Then we can get the first draft token after the first MTP module forward. For the sequential MTP modules, we can prepare their inputs in a similar way to the MTP modules in the context phase, so all of those MTP modules have the same input sequence length. After predicting all of the draft tokens, we need to evict the keys/values of those rejected draft tokens from the main model's KV cache to ensure the subsequent calculation is correct. + +### MTP Eagle + +
    +
    + tech_blog2_mtp_eagle +
    +
    +

    Figure 3. MTP Eagle, using the same notation as Figure 2

    + +MTP Eagle can be viewed as a variant of [Eagle](https://arxiv.org/pdf/2401.15077) speculative decoding method, but only supports chain decoding now. It reuses the same MTP module and repeats multiple times to predict draft tokens. MTP Eagle supports the model checkpoint with only one MTP module. The official DeepSeek-V3 and DeepSeek-R1 have only one MTP module in their checkpoints. Another difference with MTP vanilla is the KV cache. In the MTP Eagle method, the MTP module reuses the same KV cache when predicting multiple draft tokens. + +Figure 3 gives an MTP Eagle example. In the context phase, the inputs of the first MTP module forward are the same as the MTP Vanilla. However, for the sequential MTP module forward, the first difference is that MTP Eagle uses the same MTP module to predict draft tokens and reuses the same KV cache. Another difference is that we only need to input the token ID and the hidden state of one token. The token is the last predicted draft token, while the hidden state is the corresponding hidden state in the last MTP module forward. In this way, we can predict total K draft tokens by using only one MTP module. + +In the generation phase, the verification stage is the same as MTP Vanilla. After getting the accepted tokens, we will use the last accepted tokens and the corresponding hidden state as the inputs of the first MTP module forward. Compared with MTP Vanilla, it will be much easier to implement. And the sequential MTP module forwards use the same method as the context phase to prepare inputs. After predicting all of the draft tokens, we need to evict the keys/values of those rejected draft tokens from the main model's KV cache. + +## MTP implementation in TensorRT-LLM +### Basic Implementation +TensorRT-LLM has two different paths for MTP, one for [MTP Vanilla](https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/_torch/speculative/mtp.py#L1047) and another for [MTP Eagle](https://github.com/NVIDIA/TensorRT-LLM/blob/main/tensorrt_llm/_torch/speculative/mtp.py#L1047). MTP Eagle is the default path for DeepSeek-V3 and DeepSeek-R1 models. + +
    +
    + tech_blog2_overall_workflow +
    +
    +

    Figure 4. MTP workflow in TensorRT-LLM

    + +Figure 4 shows the overall workflow of MTP in TensorRT-LLM. Both paths share the runtime workflow, and the differences are in the MTP modules forward. In the context phase, there is no draft token in the inputs. TensorRT-LLM model engine fetches the input IDs from the requests and inputs to the model engine forward to get the next token and the hidden state. Then we prepare the MTP module inputs, and the MTP modules forward the inputs to predict the draft tokens. + +The generation workflow is more complicated. We need to do both the verification and draft stages. The predicted new token and draft tokens are the inputs for the main model. After the main model forward, we can sample from the output logits and get the following new tokens. Then compare them with the input draft tokens to get the final accepted tokens. The verification stage will be finished here. We will use the accepted tokens and hidden states to start a new draft stage, which uses the MTP layers to predict new draft tokens for the next iteration. Finally, we need to rewind the KV cache to evict keys/values corresponding to those rejected tokens. + +Except for the Rewind KV Cache, all of those processes are inside the model engine forward function. In this way, we can use one model engine to support MTP inference, and it would be easier for MTP to be compatible with other features, such as CUDA graph and overlap scheduler. When enabling CUDA graph, both the verification and draft stages can be captured in one graph, significantly reducing CPU overhead. + +### MTP Modules + +
    +
    + tech_blog2_mtp_modules +
    +
    +

    Figure 5. MTP model architecture

    + +Figure 5 introduces the basic model architecture of [MTP Vanilla](https://github.com/NVIDIA/TensorRT-LLM/blob/338744fba6a91147b739b7f02d19b37bc19aa17a/tensorrt_llm/_torch/speculative/mtp.py#L326), [MTP Eagle](https://github.com/NVIDIA/TensorRT-LLM/blob/338744fba6a91147b739b7f02d19b37bc19aa17a/tensorrt_llm/_torch/speculative/mtp.py#L1047), and the basic [MTP module](https://github.com/NVIDIA/TensorRT-LLM/blob/338744fba6a91147b739b7f02d19b37bc19aa17a/tensorrt_llm/_torch/models/modeling_deepseekv3.py#L829) design. Because MTP vanilla needs $K$ input tokens, if the number of accepted tokens is less than the number of input tokens, i.e. $j
    RouterGEMM#
    -

    By leveraging our internal AI code generator, we automatically generate an optimized RouterGEMM kernel, which delivers substantial improvements over the default GEMM implementation when num_tokens <=30

    +

    By leveraging our internal AI code generator, we automatically generate an optimized RouterGEMM kernel, which delivers substantial improvements over the default GEMM implementation when num_tokens <=30.

    tech_blog1_router_gemm
    @@ -1011,6 +1016,15 @@

    New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget

    + +
    +

    next

    +

    DeepSeek R1 MTP Implementation and Optimization

    +
    + +     @@ -1171,6 +1185,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.html b/latest/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.html new file mode 100644 index 0000000000..8bad27f901 --- /dev/null +++ b/latest/blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.html @@ -0,0 +1,946 @@ + + + + + + + + + + + + DeepSeek R1 MTP Implementation and Optimization — TensorRT-LLM + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    + + + + + + +
    + + + Ctrl+K +
    +     + +
    + +
    + + +
    +
    + + + +
    + +
    + + + + + +
    + +
    + +
    + +
    + +
    + + +
    + +
    + +
    +
    + +
    + + +
    + +
    + + + +
    + + +
    + +
    + +
    + +
    + + +
    + + +
    + + + + + +
    + +
    + + +
    +
    + + + + +
    + + + + + + + + + + + + +
    + + +
    + + + +
    + + +
    + +
    + +
    +
    + + +
    + + + +
    + +
    + +
    + +
    + +
    + +
    +
    + + + +
    +
    + + +
    +
    + + + +
    + +
    + + +
    +
    + +
    +
    + +
    + +
    + + +
    + +
    + + +
    +
    + + + + + +
    + +
    +

    DeepSeek R1 MTP Implementation and Optimization#

    +

    by NVIDIA TensorRT-LLM team

    +
    +

    Table of Contents#

    + +

    TensorRT-LLM achieves world-record inference performance for DeepSeek-R1 on NVIDIA Blackwell GPUs, where Multi-Token Prediction (MTP) delivers a significant speedup. In our previous blog post, we discussed the key optimizations that enable the outstanding inference latency of the DeepSeek-R1 model. This article dives deeper into the implementation and optimization of MTP in TensorRT-LLM.

    +
    +
    +

    MTP for inference#

    +

    Inspired by a previous research work, MTP is designed to help the DeepSeek-V3 training. It adds additional MTP modules at the end of the main model and uses them to predict additional tokens. In this way, MTP can extend the prediction scope to multiple future tokens at each position to achieve better model accuracy. During inference, those MTP modules can also be used for speculative decoding to improve the generation latency further. In this section, we will introduce the MTP speculative decoding algorithm for LLM inference.

    +
    +

    Background#

    +

    Speculative decoding is a popular technique for faster and cost-effective LLM inference. It’s based on the premise that generating multiple future tokens(especially for decode phase which is less compute bound) is more efficient than processing a single token. Speculative decoding techniques usually divide the process into a low-cost draft stage and a parallelized verification stage. The draft stage predicts draft tokens by using a small model or a subset of layers in the main model. And the verification stage uses the main model to determine how many of these draft tokens to accept, which is far more efficient than generating one token per iteration.

    +
    +
    + tech_blog2_verify_and_accept +
    +
    +

    Figure 1. Verification example

    +

    Figure 1 shows an example of how to verify and accept those draft tokens. Assuming there are a total of 5 draft tokens “ABCDE”, we will extend them to the input token “G”, and input a total of 6 tokens to the main model. After sampling, we can get six different expected tokens, then compare the expected tokens with the draft tokens and accept the longest prefix matched tokens. In this example, the tokens “ABC” are matched. Because “H” is predicted by the main model and the corresponding input token “C” is already accepted, “H” will also be accepted. In this way, we can accept four tokens in a single iteration. MTP also uses this method to verify and accept draft tokens. +For the draft stage in MTP, there are two different MTP methods, MTP vanilla and MTP eagle. They can be used for different inference cases.

    +
    +
    +

    MTP Vanilla#

    +
    +
    + tech_blog2_mtp_vanilla +
    +
    +

    Figure 2. MTP Vanilla, where ti is the input token, di is the predicted draft token, K is the number of MTP modules, and hin is the hidden state of the n-th MTP module. Note that h0 means the hidden states of the main model. (Disclaimer: the figures adapted from the original DeepSeek V3 tech report)

    +

    MTP Vanilla method is more similar to the MTP training, and it sequentially uses different MTP modules to predict multiple draft tokens. This method can support model checkpoints with weights of multiple different MTP modules. And each MTP module will have its own KV cache.

    +

    Figure 2 illustrates the MTP vanilla inference. In the context phase, assuming there are a total of four input tokens, we will get the output token $t_5$ and the hidden states after the main model forward. The output token will be appended to the input tokens, then we shift out the first token to get tokens from $t_2$ to $t_5$ as the input tokens of the first MTP module. The hidden states from the main model will be directly used as the input of the first MTP module to predict the first draft token. For the next several MTP modules, we will use the same method to prepare the inputs to predict the sequential draft tokens.

    +

    In the generation phase, there will be a little difference. The predicted token $t_5$ and the draft tokens will be used as inputs for the main model. After the main model forward, we will do the verification to get the accepted tokens. In this example, assuming $j$ draft tokens $d_6$~$d_{j+5}$ are accepted. Then prepare the MTP module inputs. Different from the context phase, we will prepare input IDs and hidden states of a total of $K$ tokens before the last accepted token. In this example, the last accepted token is $t_{j+6}$. Then we can get the first draft token after the first MTP module forward. For the sequential MTP modules, we can prepare their inputs in a similar way to the MTP modules in the context phase, so all of those MTP modules have the same input sequence length. After predicting all of the draft tokens, we need to evict the keys/values of those rejected draft tokens from the main model’s KV cache to ensure the subsequent calculation is correct.

    +
    +
    +

    MTP Eagle#

    +
    +
    + tech_blog2_mtp_eagle +
    +
    +

    Figure 3. MTP Eagle, using the same notation as Figure 2

    +

    MTP Eagle can be viewed as a variant of Eagle speculative decoding method, but only supports chain decoding now. It reuses the same MTP module and repeats multiple times to predict draft tokens. MTP Eagle supports the model checkpoint with only one MTP module. The official DeepSeek-V3 and DeepSeek-R1 have only one MTP module in their checkpoints. Another difference with MTP vanilla is the KV cache. In the MTP Eagle method, the MTP module reuses the same KV cache when predicting multiple draft tokens.

    +

    Figure 3 gives an MTP Eagle example. In the context phase, the inputs of the first MTP module forward are the same as the MTP Vanilla. However, for the sequential MTP module forward, the first difference is that MTP Eagle uses the same MTP module to predict draft tokens and reuses the same KV cache. Another difference is that we only need to input the token ID and the hidden state of one token. The token is the last predicted draft token, while the hidden state is the corresponding hidden state in the last MTP module forward. In this way, we can predict total K draft tokens by using only one MTP module.

    +

    In the generation phase, the verification stage is the same as MTP Vanilla. After getting the accepted tokens, we will use the last accepted tokens and the corresponding hidden state as the inputs of the first MTP module forward. Compared with MTP Vanilla, it will be much easier to implement. And the sequential MTP module forwards use the same method as the context phase to prepare inputs. After predicting all of the draft tokens, we need to evict the keys/values of those rejected draft tokens from the main model’s KV cache.

    +
    +
    +
    +

    MTP implementation in TensorRT-LLM#

    +
    +

    Basic Implementation#

    +

    TensorRT-LLM has two different paths for MTP, one for MTP Vanilla and another for MTP Eagle. MTP Eagle is the default path for DeepSeek-V3 and DeepSeek-R1 models.

    +
    +
    + tech_blog2_overall_workflow +
    +
    +

    Figure 4. MTP workflow in TensorRT-LLM

    +

    Figure 4 shows the overall workflow of MTP in TensorRT-LLM. Both paths share the runtime workflow, and the differences are in the MTP modules forward. In the context phase, there is no draft token in the inputs. TensorRT-LLM model engine fetches the input IDs from the requests and inputs to the model engine forward to get the next token and the hidden state. Then we prepare the MTP module inputs, and the MTP modules forward the inputs to predict the draft tokens.

    +

    The generation workflow is more complicated. We need to do both the verification and draft stages. The predicted new token and draft tokens are the inputs for the main model. After the main model forward, we can sample from the output logits and get the following new tokens. Then compare them with the input draft tokens to get the final accepted tokens. The verification stage will be finished here. We will use the accepted tokens and hidden states to start a new draft stage, which uses the MTP layers to predict new draft tokens for the next iteration. Finally, we need to rewind the KV cache to evict keys/values corresponding to those rejected tokens.

    +

    Except for the Rewind KV Cache, all of those processes are inside the model engine forward function. In this way, we can use one model engine to support MTP inference, and it would be easier for MTP to be compatible with other features, such as CUDA graph and overlap scheduler. When enabling CUDA graph, both the verification and draft stages can be captured in one graph, significantly reducing CPU overhead.

    +
    +
    +

    MTP Modules#

    +
    +
    + tech_blog2_mtp_modules +
    +
    +

    Figure 5. MTP model architecture

    +

    Figure 5 introduces the basic model architecture of MTP Vanilla, MTP Eagle, and the basic MTP module design. Because MTP vanilla needs $K$ input tokens, if the number of accepted tokens is less than the number of input tokens, i.e. $j<K$, we need to use the old token IDs and hidden states as the input of the first MTP module. To avoid bringing much additional computation overhead, we add two tensors for each request to save the past $K$ input IDs and the hidden states of past $K$ tokens, and update them by using the accepted tokens and corresponding hidden states each iteration. In this way, we can read these tensors when preparing inputs for the first MTP module. MTP Eagle implementation is much easier and straightforward, just call the same MTP module forward $K$ times to get $K$ new draft tokens.

    +

    The MTP module follows the design in DeepSeek-V3. The embedding layer and output head in MTP modules are shared with the main model, which can save GPU memory consumption.

    +
    +
    +

    Attention for MTP#

    +

    Attention is also a very important component in supporting MTP inference. The changes are mainly in the attention kernels for the generation phase. For the normal request, there will be only one input token in the generation phase, but for MTP, there will be $K+1$ input tokens. Since MTP sequentially predicts additional tokens, the predicted draft tokens are chained. Though we have an MTP Eagle path, currently, we only have the chain-based support for MTP Eagle. So, a causal mask is enough for the attention kernel to support MTP. In our implementation, TensorRT-LLM will use the fp8 flashMLA generation kernel on Hopper GPU, while using TRTLLM customized attention kernels on Blackwell for better performance.

    +
    +
    +

    How to run DeepSeek models with MTP#

    +

    Run DeepSeek-V3/R1 models with MTP, use examples/pytorch/quickstart_advanced.py with additional options:

    +
    cd examples/pytorch
+python quickstart_advanced.py --model_dir <YOUR_MODEL_DIR> --spec_decode_algo MTP --spec_decode_nextn N
+
    +
    +

    To benchmark min-latency performance with MTP, you need to follow this document to prepare your dataset, then follow the steps below:

    +
    YOUR_DATA_PATH=<your dataset file following the format>
+
+cat >./extra-llm-api-config.yml<<EOF
+use_cuda_graph: true
+moe_backend: TRTLLM
+speculative_config:
+    decoding_type: MTP
+    num_nextn_predict_layers: 3
+EOF
+
+export TRTLLM_ENABLE_PDL=1
+
+trtllm-bench --model nvidia/DeepSeek-R1-FP4 \
+    throughput \
+    --dataset $YOUR_DATA_PATH \
+    --backend pytorch \
+    --num_requests 10 \
+    --concurrency 1 \
+    --max_batch_size 1 \
+    --tp 8 \
+    --ep 2 \
+    --extra_llm_api_options ./extra-llm-api-config.yml
+
    +
    +
    +
    +
    +

    MTP optimization - Relaxed Acceptance#

    +

    DeepSeek-R1 is a reasoning model that first outputs some thinking tokens, after which the user can get the actual outputs. The thinking process usually takes up a lot of tokens, and the quality of the outputs of the thinking process may have a limited impact on the final answer. So we want to use a more aggressive acceptance strategy, called relaxed acceptance, for the thinking process to speed up the thinking decoding phase. This will be a tradeoff between speedup and output quality. From the experimental results, the impact of relaxed acceptance on output quality is limited.

    +
    +

    Relaxed Acceptance#

    +
    +
    + tech_blog2_relaxed_acceptance +
    +
    +

    Figure 6. Relaxed Acceptance example. Use MTP nextn=4 and top-3 in this example.

    +

    In previous verification and acceptance, we will use a top-1 to sample from the logits the main model to get the “expected” tokens as shown in Figure 1. There will be only one choice to compare with the draft tokens, which we call “Strict Acceptance”.

    +

    As for the Relaxed Acceptance, we first get the top-N tokens sampled from the logits, so more candidates will be compared with the input draft tokens. To make sure the accepted tokens are as accurate as possible, we also added a probability threshold, i.e., delta. We can get the token probabilities by applying a softmax to the logits. After getting the top-N tokens, we will remove tokens from the candidate list if their probability is smaller than the (top-1 probability - delta). In this way, we may get more than one token candidate, and all of those tokens are with a high probability. Then we can compare the input draft tokens with those candidates. If one of them matches, we can accept this draft token, so the acceptance rate will be increased. Figure 6 shows an example of a comparison between Strict Acceptance and Relaxed Acceptance.

    +

    Note that the Relaxed Acceptance will only be used during the thinking phase, while the Strict Acceptance will still be used during the non-thinking phase. And the Relaxed Acceptance only supports the DeepSeek-R1 model now.

    +
    +
    +

    How to run the DeepSeek-R1 model with Relaxed Acceptance#

    +

    Run DeepSeek-R1 models with MTP Relaxed Acceptance, use examples/pytorch/quickstart_advanced.py with additional options:

    +
    cd examples/pytorch
+python quickstart_advanced.py --model_dir <YOUR_MODEL_DIR> --spec_decode_algo MTP --spec_decode_nextn N --use_relaxed_acceptance_for_thinking --relaxed_topk 10 --relaxed_delta 0.6
+
    +
    +

    To benchmark min-latency performance with MTP Relaxed Acceptance, you need to follow this document to prepare your dataset, then follow the steps below:

    +
    YOUR_DATA_PATH=<your dataset file following the format>
+
+cat >./extra-llm-api-config.yml<<EOF
+use_cuda_graph: true
+moe_backend: TRTLLM
+speculative_config:
+    decoding_type: MTP
+    num_nextn_predict_layers: 3
+    use_relaxed_acceptance_for_thinking: true
+    relaxed_topk: 10
+    relaxed_delta: 0.6
+EOF
+
+export TRTLLM_ENABLE_PDL=1
+
+trtllm-bench --model nvidia/DeepSeek-R1-FP4 \
+    throughput \
+    --dataset $YOUR_DATA_PATH \
+    --backend pytorch \
+    --num_requests 10 \
+    --concurrency 1 \
+    --max_batch_size 1 \
+    --tp 8 \
+    --ep 2 \
+    --extra_llm_api_options ./extra-llm-api-config.yml
+
    +
    +
    +
    +
    +

    Evaluation#

    +
    +

    Achieving speedup with MTP speculative decoding#

    +
    +
    + tech_blog2_perf_and_ar +
    +
    +

    Figure 7. DeepSeek-R1-FP4 671B min-latency performance with different MTP next-n

    +

    We tested the min-latency (batch size = 1) performance of the DeepSeek-R1-FP4 model with different MTP next-n on a B200 node. The MLA runs with TP=8, and the MoE runs with EP=2. And there are ten different requests with ISL/OSL=1K/2K. From Figure 7, we can see that MTP=3 can help get the best min-latency performance on 8 B200 GPUs, which can bring 2.16x speedup compared with the baseline nextn=0. And with the help of the relaxed acceptance, the min-latency performance can be further improved to achieve a 2.33x speedup. We also evaluated the CUDA graph and overlap scheduler benefits. For such a min-latency case, CUDA graph can achieve a 7.22x average speedup, while the overlap scheduler can achieve 1.03x average latency.

    +
    +
    +

    Accuracy studies for Relaxed Acceptance#

    +
    +
    + tech_blog2_acc_relaxed_acceptance +
    +
    +

    Figure 8. Ablation results for the Relaxed Acceptance. Using MTP nextn=3, top-10, and delta=0.6.

    +

    We validated the Relaxed Acceptance on different datasets. In Figure 8, we show the ablation results for Relaxed Acceptance by using the DeepSeek-R1-FP4 model. Compared with Strict Acceptance, the impact of Relaxed Acceptance on output quality is limited, resulting in only a slight accuracy drop.

    +
    +
    +
    +

    Future Works#

    +
    +

    Tree-based speculative decoding support#

    +
    +
    + tech_blog2_tree_spec_decoding +
    +
    +

    Figure 9. Comparison between the chain-based and tree-based speculative decoding

    +

    TensorRT-LLM PyTorch backend can only support chain-based speculative decoding now, both MTP Vanilla and MTP Eagle. However, the tree-based speculative decoding technique is widely used in previous advanced methods, such as Ealge2 and Eagle3, to increase the acceptance rate. MTPs in TensorRT-LLM can also be extended to support the tree-based technique. Figure 9 compares the chain-based method with the tree-based method. Both full tree and dynamic tree methods can help expand the candidate combinations, so that we can have more choices for the draft tokens.

    +
    +
    +

    Eagle3 support#

    +

    Another important method is Eagle3. From the Eagle3 paper, the promising results show that it can help greatly increase the acceptance rate by leveraging different levels’ hidden states to predict draft tokens. Since TensorRT-LLM already has Eagle-3 support now, in the future, we also want to train an Eagle3 head to support DeepSeek-V3/R1+Eagle3 to achieve better speedup.

    +
    +
    +

    Fix known issues#

    +

    There are still some known issues, and we will fix them soon:

    +
      +

    • The MTP vanilla path has a known accuracy issue. We will fix it and refactor the MTP vanilla implementation.

      • +

    • The MTP Eagle is non-deterministic now.

      • +

    • An accuracy issue when enabling MTP and attention DP together.

      • +
    +
    +
    +
    +

    Acknowledgment#

    +

    This was a remarkable cross-team effort to support and optimize MTP in TensorRT-LLM. We would like to extend our gratitude to everyone who contributed to making this possible, as it involved a typical system/algorithm co-design approach spanning multiple technical layers—including kernel optimization, runtime enhancements, algorithmic improvements, and performance measurement & analysis. And a special thanks goes to the DeepSeek team for developing the MTP method, which lays down the foundation of this blog.

    +
    +
    + + +
    + + + + + + + +
    + + + + + + + + + + +
    +
    + +
    + +
    +
    +
    + + + + + + + + \ No newline at end of file diff --git a/latest/blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.html b/latest/blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.html new file mode 100644 index 0000000000..116bb0c4cb --- /dev/null +++ b/latest/blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.html @@ -0,0 +1,904 @@ + + + + + + + + + + + + Optimizing DeepSeek R1 Throughput on NVIDIA Blackwell GPUs: A Deep Dive for Developers — TensorRT-LLM + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    + + + + + + +
    + + + Ctrl+K +
    +     + +
    + +
    + + +
    +
    + + + +
    + +
    + + + + + +
    + +
    + +
    + +
    + +
    + + +
    + +
    + +
    +
    + +
    + + +
    + +
    + + + +
    + + +
    + +
    + +
    + +
    + + +
    + + +
    + + + + + +
    + +
    + + +
    +
    + + + + +
    + + + + + + + + + + + + +
    + + +
    + + + +
    + + +
    + +
    + +
    +
    + + +
    + + + +
    + +
    + +
    + +
    + +
    + +
    +
    + + + +
    +
    + + +
    +
    + + + +
    + +
    + + +
    +
    + +
    +
    + +
    + +
    + + +
    + +
    + + +
    +
    + + + + + +
    + +
    +

    Optimizing DeepSeek R1 Throughput on NVIDIA Blackwell GPUs: A Deep Dive for Developers#

    +

    By NVIDIA TensorRT-LLM team

    +
    +

    Table of Contents#

    + +
    +
    +

    Introduction#

    +

    The open source DeepSeek R1 model’s innovative architecture including the multi-head latent attention (MLA) and large sparse Mixture-of-Experts (MoE) significantly improved the inference efficiency of the LLM models. However, harnessing the full potential of such an innovative structure requires equally important hardware/software co-optimization. This post delves into the optimization strategies for DeepSeek R1 throughput oriented scenarios (TPS/GPU), developed by NVIDIA within TensorRT-LLM on NVIDIA’s Blackwell B200 GPUs. We will explore the rationale behind each enhancement. The other min-latency optimization blog explained in detail how TensorRT-LLM optimizes the R1 performance to achieve the best of the TPS/USER.

    +

    These optimizations have significantly boosted DeepSeek R1 throughput on Blackwell. Performance increased from approximately 2000 TPS/GPU in February to 4600 TPS/GPU on ISL/OSL 1K/2K dataset. The optimizations are general and applicable to other ISL/OSL configs too. These optimization items were broadly categorized into three areas: MLA layers, MoE layers, and runtime.

    +
    +
    +

    Precision strategy#

    +

    The mixed precision recipe for DeepSeek R1 throughput scenario is almost the same as what is used for latency oriented scenario, with the following differences:

    +
      +

    • FP8 KV cache and FP8 attention, rather than BF16 precision.

      • +

    • FP4 Allgather for better communication bandwidth utilization.

      • +
    +

    The checkpoint used in this blog is hosted in nvidia/DeepSeek-R1-FP4, generated by NVIDIA Model Optimizer. The accuracy score of common dataset on this FP4 checkpoint and TensorRT-LLM implementations are:

    +
    + + + + + + + + + + + + + + + + +

    Precision

    GPQA Diamond

    MATH-500

    TensorRT-LLM FP8

    0.697

    0.954

    TensorRT-LLM FP4

    0.705

    0.96

    +
    +

    ** Note there are some run-to-run variance for these evaluations, so FP4 data is slight higher here. We think FP4 has comparable accuracy with FP8 on these datasets.

    +

    The MoE layers inside this checkpoint have been quantized into FP4. Quantizing the MoE layer weights into FP4 has the following benefits:

    +
      +

    • Fully utilize the 5th generation Tensor Core FLOPS of the NVIDIA Blackwell GPUs

      • +

    • Reduce the memory load needs of the weights by almost half for MoE. Since the MoE parts are still memory bound for the decoding phase for the scenario, and 97% of the weights in the DeepSeek R1 model are from MoE layers.

      • +

    • Reduce the memory footprint of the model weights, thus freeing more GPU memories for KV cache and then increasing the max concurrency. The original FP8 model checkpoint of the DeepSeek R1 model is about 640GB, while the NVIDIA provided DeepSeek R1 FP4 quantized model is only about 400 GB.

      • +
    +

    The precision of FP8 KV cache and FP8 attention kernels are evaluated on the GSM8K dataset, with no obvious accuracy drops. For the accuracy numbers, please see the table in the FP8 KV cache section. Users can still opt-out to use BF16 KV cache and attention if on their dataset some accuracy differences are observed.

    +
    +
    +

    Parallel strategy#

    +

    The parallelism strategy for DeepSeek R1 throughput scenario is different from what is used for latency-oriented scenarios.

    +
    + + + + + + + + + + + + + + + + + + + + + + +

    Components

    Parallel Patterns

    Attention Modules

    Data Parallelism 8 (DP8)

    MoE Sparse Experts

    Expert Parallelism 8 (EP8)

    MoE Shared Experts

    DP8

    Fuse_A GEMM

    DP8

    Router GEMM

    DP8

    +
    +

    In the following sections we will explain the rationale why DP and EP are chosen and not using tensor parallel (TP).

    +
    +

    Weights absorb and MQA#

    +

    The core idea of MLA is the low-rank joint compression for the attention keys and values to reduce KV-cache size during the inference. Based on the MLA formulas, the down-projected KV latent is up-projected to multiple heads and combined with the up-projected Q to establish a normal multi-head attention (MHA). Due to the nature of the matrix multiplication, the up projection weights matrix of the K (W^UK) can be multiplied by the up-projection weights matrix of Q (W^Q) firstly, the computed results of these 2 can be then multiplied to Q. The up-projection weights matrix of V (W^UV) and the attention output projection matrix W^O can also be multiplied after the attention output. The DeepSeek-V2 technical report calls this technique “absorb”. After the weights are absorbed, the MLA is equivalent to multiple query attention(MQA). Please see the original DeepSeek-V2 technical paper for the detailed formulas and explanations, the following block diagram shows the computational flow of weights absorbed MLA in TensorRT-LLM. +Weights Absorb

    +

    For the decoding phase, the weights absorb significantly reduces the math FLOPS needed to up project the K and V, since the FLOPs needed for these up projections of KV are linear to the KV cache length, while length of Q vector is always 1 in the decoding phase. The longer the KV cache history is, the more FLOPs are needed, and the up projections are repeated for every decoded token since only the projected KV latent were saved, which further increases the FLOPs needed. +For the prefill phase, the weights absorbed version changes the dimensions of Q and KV thus increasing the number of FLOPs for attention. Based on roofline analysis, non absorbed version is beneficial for the prefill phase with input length 256 or larger +The TensorRT-LLM MLA implementation chooses different highly optimized kernels for prefill and decoding, see MLA.

    +
    +
    +

    Data Parallel for Attention module (ADP)#

    +

    The intuition of choosing attention DP is that doing TP for the MQA (where different GPUs compute different attention Q heads) will duplicate the KV cache memory, which limits the concurrency being achieved by the system. The duplication factor is equal to the TP group size, thus 8x for TP8. Small concurrency will hurt the throughput for the powerful system like NVIDIA DGX B200.

    +

    For DeepSeek R1 FP4 checkpoint with 8 B200 GPUs, the weights and activation occupies about 80 GB memory for each GPU, and the free KV cache per GPU will be 100GB. Assuming ISL 1K, OSL 2K, each request will consume about 200MB KV cache, which results in a per GPU max concurrency of 500. A single node 8xGPU system has a global concurrency of 4000. When using attention TP, the global concurrency will become just 500.

    +

    Silicon experiments show the attention DP technique provides a significant 400% speedup in the max throughput cases, when keeping all other factors the same.

    +
    +
    +

    Expert parallel for MoE (EP)#

    +

    The DeepSeek R1 MoE design features 256 small sparse experts and 1 shared expert, the GEMM problem size of these experts are as follows.

    +
    + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

    GEMM

    group

    GEMM N

    GEMM K

    shared_fc1

    1

    4096

    7168

    shared_fc2

    1

    7168

    2048

    sparse_fc1

    256

    4096

    7168

    sparse_fc2

    256

    7168

    2048

    +
    +

    These experts can be done in either Tensor-Parallelism or Expert-Parallelism ways. Our current ablation study reveals that Expert-Parallelism achieves better GEMM FLOPS because it has better GEMM problem sizes. And Expert-Parallelism can save GPU communication bandwidth compared to AllReduce, because the tokens only need to be sent to GPUs where the active experts for this token are located, while TP needs an AllReduce for all the tokens between all the GPUs. Also to be noted that, to scale the DeepSeek R1 inference to systems like GB200 NVL72 fully utilizing the aggregated memory bandwidth and tensor core flops, large EPs are needed. We are actively working on implementing it.

    +

    Silicon performance measurements show that Expert-Parallelism can provide 142% speedup for 1K/2K max throughput case, when keeping other factors the same.

    +
    +
    +
    +

    MLA Layers Optimizations#

    +

    Other than the parallel strategy and precision strategy we explained above, we have done the following optimizations for layers/kernels inside the MLA module.

    +
      +

    • Attention Kernels Optimization

      +

      This provided a 20% E2E speedup compared to February baseline implementation. It involved implementing high-throughput generation MLA kernels. Techniques include using 2CTA Group variant of the Tensor Core 5th MMA instructions of Blackwell GPUs, overlapping MLA with softmax using interleaved tiles, and fine-tuning kernel selection heuristics for the DeepSeek R1 problem size.

      +
      • +

    • FP8 KV Cache

      +

      An important optimization that yielded a 6% E2E throughput increase when assuming the concurrency was identical. Another benefit of FP8 KV cache is compressing the KV cache size by half, which allows for larger concurrency. It also enables the use of faster FP8 attention kernels compared to BF16. We recommend that users always turn on FP8 KV cache to get better performance. In the context phase, KV is quantized to FP8 and saved to the KV cache pool. In the generation phase, both Q and KV are quantized to FP8, and FP8 Multi-Query Attention (MQA) is used. Evaluation on GSM8k showed no meaningful accuracy drop. The quantization typically uses static per-tensor FP8 with a scaling factor defaulting to 1.0, but KV cache scaling factor can also be generated by calibrating on a target dataset. Below are the accuracy metrics of different combinations on the GSM8K dataset.

      +
      + + + + + + + + + + + + + + + + +

      KV Cache Type

      FP8 Checkpoint

      FP4 Checkpoint

      BF16 MLA and KV cache

      0.9629

      0.9606

      FP8 MLA and KV cache

      0.9613

      0.9606

      +
      +
      • +

    • Manual GEMM tactics tuning

      +

      This optimization addresses cases where the default heuristic algorithm in cuBLAS is not performing best for specific GEMM shapes existing in the model. We built an internal tool to find the best algorithm for these specific shapes offline and then used the cublasLtMatmul API to apply this specific, optimized algorithm at runtime. This is a necessary system optimization when general-purpose heuristics don’t find the most efficient kernel for all specific cases. We are also working actively with the cuBLAS team to further enhance the heuristics such that the best performance can always be achieved OOTB. See cublasScaledMM.cpp for the tuning details.

      +
      • +

    • Horizontal Fusions

      +

      This involves fusing GEMM operations of down projection of Q/KV and rope dimensions of K tensor. See modeling_deepseekv3.py for details. Horizontal fusion reduces the kernel launch overhead and increases the GEMM problem sizes which can achieve better HW utilization. It is a common technique shared by both min-latency and throughput optimizations.

      +
      • +

    • 2-stream optimizations

      +

      There are some small operations which can be run in parallel like the Q norm and KV norm inside the MLA. These operations cannot fully utilize the GPU math flops and the memory bandwidth, thus running in parallel CUDA streams can bring speed-up.

      +
      • +
    +
    +
    +

    MoE Layers Optimizations#

    +

    The following optimizations are already done for MoE layers.

    +
      +

    • Mix I/O data type for the router GEMM

      +

      Achieved a 4% E2E speedup by avoiding casting operations and performing the GEMM using a mixture of input and output data types (e.g., BF16 input and FP32 output) directly. This eliminates the need to explicitly cast inputs to the output type and saves memory bandwidth.

      +
      • +

    • Top-K Kernels Fusions

      +

      Resulted in a 7.4% E2E speedup. For DeepSeek R1, selecting the top 8 experts from 256 is done in a two-phase approach: first selecting top groups, then finding the top 8 within those groups. DeepSeek R1 uses some additional techniques for better expert load balance which involves adding bias and scales to the topK complications. All these operations resulted in 18 PyTorch ops when not fused, see Deepseekv3RoutingImpl. Fusing the multiple kernels involved in these Top-K calculations significantly reduces the overall computation time. Compared to using 18 native PyTorch ops, fusion can reduce the operation to as few as 2 kernels. Based on the measurement on B200, fusing these kernels can reduce the kernel time from 252us to 15us in the target setting.

      +
      • +

    • FP4 AllGather Optimizations

      +

      Showed a 4% E2E speedup. This optimization replaces the BF16 AllGather operation with an FP4 version. Using a lower precision for this communication primitive reduces the amount of data transferred over the network, significantly improving communication efficiency. Also, since the original BF16 Tensor to be transferred will get cast into FP4 format after the AllGather communication, this optimization will not bring any impact to the accuracy. At the kernel level, we are seeing about 3x when switching from BF16 to FP4 AllGather.

      +
      • +

    • CUTLASS Group GEMM optimizations

      +

      Provided a 1.3% E2E speedup. There are some CUTLASS level optimizations shared by both min-latency and throughput cases. Just updating CUTLASS to the latest version gives us 13% kernel improvement for the MoE groupGemm, and resulted in +1.3% E2E TPS/GPU.

      +
      • +

    • Multi-stream optimizations +Running the shared and routed experts in 2 streams combined with other multi-streaming optimizations in the MLA modules, contributing a 5.3% E2E speedup.

      • +
    +
    +
    +

    Runtime Optimizations#

    +

    These optimizations target the overall execution flow, scheduling, and resource management within the inference system. They are shared between DeepSeek R1 models and other models supported in the TensorRT-LLM, here we are sharing some ablation study for the performance benefits on DeepSeek R1 on B200.

    +
      +

    • CUDA Graph

      +

      This had a significant 22% E2E performance impact for throughput scenarios. CUDA Graphs allow capturing a sequence of CUDA operations and launching them as a single unit, drastically reducing kernel launch overheads. This is particularly beneficial for models with many small kernels, and particularly on the PyTorch flow, because the python host code normally executes slower than C++. Since the CUDA Graph freezes the kernel launch parameters, which is normally associated with the tensor shapes, it can only be safely used with static shape, meaning that different CUDA graphs need to be captured for different batch sizes. Each graph will have some cost of memory usage, and capturing time, thus we cannot capture every possible CUDA graph for all possible batches. For the non-captured batch sizes, PyTorch eager mode code will be executed. There is a feature called CUDA Graph padding in TensorRT-LLM, which is a good trade-off between the number of CUDA Graphs and the CUDA Graph hit ratio; it tries to pad a batch to the nearest one with a captured CUDA Graph. Normally you should enable the CUDA Graph padding feature to increase the CUDA Graph hit rate, but the padding itself has some overhead due to wasted tokens computation. Users can opt-out the CUDA Graph padding feature to see the perf benefits, by setting the cuda_graph_padding_enabled to false, see API here Pytorch backend config

      +
      • +

    • Overlap Scheduler:

      +

      Showed a 4% E2E performance impact and should generally always be used. This scheduler manages the execution of different operations (like computation and communication) to overlap them effectively on the GPU and network. The intuition is to hide latency by performing computation while waiting for data transfers or vice versa, improving overall hardware utilization. The overlap schedule is already defaulted on in TensorRT-LLM by commit. In case there are corner cases where it does not work, users can still opt-out this feature by set disable_overlap_scheduler to true.

      +
      • +

    • Memory Optimizations

      +

      Resulted in a 4GB improvement. This includes techniques like chunked MoE (specifically for Hopper) and fixing a cuda context init bug. These methods reduce the memory footprint of the model weights or intermediate tensors, allowing for larger batch sizes or sequence lengths, and preventing Out-of-Memory (OOM) errors.

      +
      • +
    +
    +
    +

    How to reproduce#

    +

    See Perf practices

    +
    +
    +

    Future Works#

    +
      +

    • Large EP

      • +

    • Chunked context

      • +

    • More communication overlap

      • +
    +
    +
    +

    Acknowledgment#

    +

    The substantial throughput advancements for DeepSeek R1 on Blackwell GPUs, as detailed in this post, are the fruit of a dedicated and collaborative engineering effort. Achieving nearly a 2.3x increase in TPS/GPU required a deep dive into MLA layers, MoE layers, and runtime optimizations. We extend our sincere appreciation to all the engineers involved in this intensive optimization process. Their collective expertise in pushing the boundaries of throughput performance within TensorRT-LLM has been instrumental. We trust that sharing these specific strategies for maximizing throughput will prove beneficial to the developer community as they tackle demanding LLM inference workloads on NVIDIA hardware.

    +
    +
    + + +
    + + + + + +
    + +
    +
    +
    + +
    + + + + + + + + + + +
    +
    + +
    + +
    +
    +
    + + + + + + + + \ No newline at end of file diff --git a/latest/commands/trtllm-build.html b/latest/commands/trtllm-build.html index f49eeb6e73..7748f9223a 100644 --- a/latest/commands/trtllm-build.html +++ b/latest/commands/trtllm-build.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -357,6 +358,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -421,6 +423,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -455,6 +458,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -529,7 +533,7 @@ [--profiling_verbosity {layer_names_only,detailed,none}] [--strip_plan] [--weight_sparsity] [--weight_streaming] [--fast_build] [--workers WORKERS] - [--log_level {internal_error,error,warning,info,verbose,debug}] + [--log_level {internal_error,error,warning,info,verbose,debug,trace}] [--enable_debug_output] [--visualize_network VISUALIZE_NETWORK] [--dry_run] [--monitor_memory] [--logits_dtype {float16,float32}] @@ -665,7 +669,7 @@

    Default: 1

    --log_level
    -

    Possible choices: internal_error, error, warning, info, verbose, debug

    +

    Possible choices: internal_error, error, warning, info, verbose, debug, trace

    The logging level.

    Default: 'info'

    @@ -1049,6 +1053,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/commands/trtllm-serve.html b/latest/commands/trtllm-serve.html index 9e7730ba21..681eef74b6 100644 --- a/latest/commands/trtllm-serve.html +++ b/latest/commands/trtllm-serve.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -357,6 +358,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -421,6 +423,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -455,6 +458,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -778,7 +782,7 @@ However, for the PyTorch backend, specified with the
    Options:
    -

    internal_error | error | warning | info | verbose | debug

    +

    internal_error | error | warning | info | verbose | debug | trace

    @@ -827,7 +831,7 @@ However, for the PyTorch backend, specified with the
    Options:
    -

    internal_error | error | warning | info | verbose | debug

    +

    internal_error | error | warning | info | verbose | debug | trace

    @@ -1094,6 +1098,15 @@ However, for the PyTorch backend, specified with the +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/dev-on-cloud/build-image-to-dockerhub.html b/latest/dev-on-cloud/build-image-to-dockerhub.html index e72ebdafa9..1525aa566f 100644 --- a/latest/dev-on-cloud/build-image-to-dockerhub.html +++ b/latest/dev-on-cloud/build-image-to-dockerhub.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -355,6 +356,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -419,6 +421,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -453,6 +456,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -678,6 +682,15 @@ docker push <your_dockerhub_use

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/dev-on-cloud/dev-on-runpod.html b/latest/dev-on-cloud/dev-on-runpod.html index d181b43aef..a35869e2b4 100644 --- a/latest/dev-on-cloud/dev-on-runpod.html +++ b/latest/dev-on-cloud/dev-on-runpod.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -355,6 +356,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -419,6 +421,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -453,6 +456,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -678,6 +682,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/curl_chat_client.html b/latest/examples/curl_chat_client.html index b5c5eafff8..e426b4d589 100644 --- a/latest/examples/curl_chat_client.html +++ b/latest/examples/curl_chat_client.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -649,6 +653,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/curl_chat_client_for_multimodal.html b/latest/examples/curl_chat_client_for_multimodal.html index d088b1ccae..301ce4394e 100644 --- a/latest/examples/curl_chat_client_for_multimodal.html +++ b/latest/examples/curl_chat_client_for_multimodal.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -726,6 +730,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/curl_completion_client.html b/latest/examples/curl_completion_client.html index 8b84daedd3..3025854bc1 100644 --- a/latest/examples/curl_completion_client.html +++ b/latest/examples/curl_completion_client.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -648,6 +652,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/customization.html b/latest/examples/customization.html index 454eb74358..669753a663 100644 --- a/latest/examples/customization.html +++ b/latest/examples/customization.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -357,6 +358,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -421,6 +423,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -455,6 +458,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -785,6 +789,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/deepseek_r1_reasoning_parser.html b/latest/examples/deepseek_r1_reasoning_parser.html index eee42aeeb5..e02ebada65 100644 --- a/latest/examples/deepseek_r1_reasoning_parser.html +++ b/latest/examples/deepseek_r1_reasoning_parser.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -649,6 +653,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/genai_perf_client.html b/latest/examples/genai_perf_client.html index 22a70a0cbb..fde5233ecc 100644 --- a/latest/examples/genai_perf_client.html +++ b/latest/examples/genai_perf_client.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -654,6 +658,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/genai_perf_client_for_multimodal.html b/latest/examples/genai_perf_client_for_multimodal.html index 89d6abe45b..d6aac1f037 100644 --- a/latest/examples/genai_perf_client_for_multimodal.html +++ b/latest/examples/genai_perf_client_for_multimodal.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -657,6 +661,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/index.html b/latest/examples/index.html index 570ccd9ac7..f4b5146a1e 100644 --- a/latest/examples/index.html +++ b/latest/examples/index.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -544,6 +548,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -689,6 +694,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_api_examples.html b/latest/examples/llm_api_examples.html index c35077212c..9e2224f70c 100644 --- a/latest/examples/llm_api_examples.html +++ b/latest/examples/llm_api_examples.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -513,6 +517,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -653,6 +658,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_auto_parallel.html b/latest/examples/llm_auto_parallel.html index 422d3f0b52..d23b5bcbbd 100644 --- a/latest/examples/llm_auto_parallel.html +++ b/latest/examples/llm_auto_parallel.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -672,6 +676,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_eagle2_decoding.html b/latest/examples/llm_eagle2_decoding.html new file mode 100644 index 0000000000..d00948ce13 --- /dev/null +++ b/latest/examples/llm_eagle2_decoding.html @@ -0,0 +1,717 @@ + + + + + + + + + + + + Generate Text Using Eagle2 Decoding — TensorRT-LLM + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
    + + + + + + +
    + + + Ctrl+K +
    +     + +
    + +
    + + +
    +
    + + + +
    + +
    + + + + + +
    + +
    + +
    + +
    + +
    + + +
    + +
    + +
    +
    + +
    + + +
    + +
    + + + +
    + + +
    + +
    + +
    + +
    + + +
    + + +
    + + + +
    + +
    + + +
    +
    + + + + +
    + + + + + + + + + + + + +
    + + +
    + + + +
    + + +
    + +
    + +
    +
    + + +
    + + + +
    + +
    + +
    + +
    + +
    + +
    +
    + + + +
    +
    + + +
    +
    + + + +
    + +
    + + +
    +
    + +
    +
    + +
    + +
    + + +
    + +
    + + +
    +
    + + + + + +
    + +
    +

    Generate Text Using Eagle2 Decoding#

    +

    Source NVIDIA/TensorRT-LLM.

    +
     1### Generate Text Using Eagle2 Decoding
+ 2
+ 3from tensorrt_llm import LLM, SamplingParams
+ 4from tensorrt_llm.llmapi import (LLM, EagleDecodingConfig, KvCacheConfig,
+ 5                                 SamplingParams)
+ 6
+ 7
+ 8def main():
+ 9    # Sample prompts.
+10    prompts = [
+11        "Hello, my name is",
+12        "The president of the United States is",
+13        "The capital of France is",
+14        "The future of AI is",
+15    ]
+16    # The end user can customize the sampling configuration with the SamplingParams class
+17    sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
+18
+19    # The end user can customize the kv cache configuration with the KVCache class
+20    kv_cache_config = KvCacheConfig(enable_block_reuse=True)
+21
+22    llm_kwargs = {}
+23
+24    model = "lmsys/vicuna-7b-v1.3"
+25
+26    # The end user can customize the eagle decoding configuration by specifying the
+27    # speculative_model, max_draft_len, num_eagle_layers, max_non_leaves_per_layer, eagle_choices
+28    # greedy_sampling,posterior_threshold, use_dynamic_tree and dynamic_tree_max_topK
+29    # with the EagleDecodingConfig class
+30
+31    speculative_config = EagleDecodingConfig(
+32        speculative_model="yuhuili/EAGLE-Vicuna-7B-v1.3",
+33        max_draft_len=63,
+34        num_eagle_layers=4,
+35        max_non_leaves_per_layer=10,
+36        use_dynamic_tree=True,
+37        dynamic_tree_max_topK=10)
+38
+39    llm = LLM(model=model,
+40              kv_cache_config=kv_cache_config,
+41              speculative_config=speculative_config,
+42              max_batch_size=1,
+43              max_seq_len=1024,
+44              **llm_kwargs)
+45
+46    outputs = llm.generate(prompts, sampling_params)
+47
+48    # Print the outputs.
+49    for output in outputs:
+50        prompt = output.prompt
+51        generated_text = output.outputs[0].text
+52        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
+53
+54
+55if __name__ == '__main__':
+56    main()
+
    +
    +
    + + +
    + + + + + + + +
    + + + +
    + + + + + +
    +
    + +
    + +
    +
    +
    + + + + + + + + \ No newline at end of file diff --git a/latest/examples/llm_eagle_decoding.html b/latest/examples/llm_eagle_decoding.html index 05d1ac4e95..003714d8bf 100644 --- a/latest/examples/llm_eagle_decoding.html +++ b/latest/examples/llm_eagle_decoding.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -511,8 +515,8 @@ 
     1### Generate Text Using Eagle Decoding
  2
  3from tensorrt_llm import LLM, SamplingParams
- 4from tensorrt_llm.llmapi import (LLM, BuildConfig, EagleDecodingConfig,
- 5                                 KvCacheConfig, SamplingParams)
+ 4from tensorrt_llm.llmapi import (LLM, EagleDecodingConfig, KvCacheConfig,
+ 5                                 SamplingParams)
  6
  7
  8def main():
@@ -526,51 +530,49 @@
 16    # The end user can customize the sampling configuration with the SamplingParams class
 17    sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
 18
-19    # The end user can customize the build configuration with the BuildConfig class
-20    build_config = BuildConfig(max_batch_size=1, max_seq_len=1024)
+19    # The end user can customize the kv cache configuration with the KVCache class
+20    kv_cache_config = KvCacheConfig(enable_block_reuse=True)
 21
-22    # The end user can customize the kv cache configuration with the KVCache class
-23    kv_cache_config = KvCacheConfig(enable_block_reuse=True)
-24
-25    llm_kwargs = {}
-26
-27    model = "lmsys/vicuna-7b-v1.3"
-28
-29    # The end user can customize the eagle decoding configuration by specifying the
-30    # speculative_model, max_draft_len, num_eagle_layers, max_non_leaves_per_layer, eagle_choices
-31    # greedy_sampling,posterior_threshold, use_dynamic_tree and dynamic_tree_max_topK
-32    # with the EagleDecodingConfig class
-33
-34    speculative_config = EagleDecodingConfig(
-35        speculative_model="yuhuili/EAGLE-Vicuna-7B-v1.3",
-36        max_draft_len=63,
-37        num_eagle_layers=4,
-38        max_non_leaves_per_layer=10,
-39                            eagle_choices=[[0], [0, 0], [1], [0, 1], [2], [0, 0, 0], [1, 0], [0, 2], [3], [0, 3], [4], [0, 4], [2, 0], \
-40                                            [0, 5], [0, 0, 1], [5], [0, 6], [6], [0, 7], [0, 1, 0], [1, 1], [7], [0, 8], [0, 0, 2], [3, 0], \
-41                                            [0, 9], [8], [9], [1, 0, 0], [0, 2, 0], [1, 2], [0, 0, 3], [4, 0], [2, 1], [0, 0, 4], [0, 0, 5], \
-42                                            [0, 0, 0, 0], [0, 1, 1], [0, 0, 6], [0, 3, 0], [5, 0], [1, 3], [0, 0, 7], [0, 0, 8], [0, 0, 9], \
-43                                            [6, 0], [0, 4, 0], [1, 4], [7, 0], [0, 1, 2], [2, 0, 0], [3, 1], [2, 2], [8, 0], \
-44                                            [0, 5, 0], [1, 5], [1, 0, 1], [0, 2, 1], [9, 0], [0, 6, 0], [0, 0, 0, 1], [1, 6], [0, 7, 0]]
-45    )
-46
-47    llm = LLM(model=model,
-48              build_config=build_config,
-49              kv_cache_config=kv_cache_config,
-50              speculative_config=speculative_config,
-51              **llm_kwargs)
+22    llm_kwargs = {}
+23
+24    model = "lmsys/vicuna-7b-v1.3"
+25
+26    # The end user can customize the eagle decoding configuration by specifying the
+27    # speculative_model, max_draft_len, num_eagle_layers, max_non_leaves_per_layer, eagle_choices
+28    # greedy_sampling,posterior_threshold, use_dynamic_tree and dynamic_tree_max_topK
+29    # with the EagleDecodingConfig class
+30
+31    speculative_config = EagleDecodingConfig(
+32        speculative_model="yuhuili/EAGLE-Vicuna-7B-v1.3",
+33        max_draft_len=63,
+34        num_eagle_layers=4,
+35        max_non_leaves_per_layer=10,
+36                            eagle_choices=[[0], [0, 0], [1], [0, 1], [2], [0, 0, 0], [1, 0], [0, 2], [3], [0, 3], [4], [0, 4], [2, 0], \
+37                                            [0, 5], [0, 0, 1], [5], [0, 6], [6], [0, 7], [0, 1, 0], [1, 1], [7], [0, 8], [0, 0, 2], [3, 0], \
+38                                            [0, 9], [8], [9], [1, 0, 0], [0, 2, 0], [1, 2], [0, 0, 3], [4, 0], [2, 1], [0, 0, 4], [0, 0, 5], \
+39                                            [0, 0, 0, 0], [0, 1, 1], [0, 0, 6], [0, 3, 0], [5, 0], [1, 3], [0, 0, 7], [0, 0, 8], [0, 0, 9], \
+40                                            [6, 0], [0, 4, 0], [1, 4], [7, 0], [0, 1, 2], [2, 0, 0], [3, 1], [2, 2], [8, 0], \
+41                                            [0, 5, 0], [1, 5], [1, 0, 1], [0, 2, 1], [9, 0], [0, 6, 0], [0, 0, 0, 1], [1, 6], [0, 7, 0]]
+42    )
+43
+44    llm = LLM(model=model,
+45              kv_cache_config=kv_cache_config,
+46              speculative_config=speculative_config,
+47              max_batch_size=1,
+48              max_seq_len=1024,
+49              **llm_kwargs)
+50
+51    outputs = llm.generate(prompts, sampling_params)
 52
-53    outputs = llm.generate(prompts, sampling_params)
-54
-55    # Print the outputs.
-56    for output in outputs:
-57        prompt = output.prompt
-58        generated_text = output.outputs[0].text
-59        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
-60
-61
-62if __name__ == '__main__':
-63    main()
+53    # Print the outputs.
+54    for output in outputs:
+55        prompt = output.prompt
+56        generated_text = output.outputs[0].text
+57        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
+58
+59
+60if __name__ == '__main__':
+61    main()
    @@ -700,6 +702,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_guided_decoding.html b/latest/examples/llm_guided_decoding.html index a26dba34ca..16eae7f081 100644 --- a/latest/examples/llm_guided_decoding.html +++ b/latest/examples/llm_guided_decoding.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -682,6 +686,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference.html b/latest/examples/llm_inference.html index d9f141cdd6..b4cd6c2808 100644 --- a/latest/examples/llm_inference.html +++ b/latest/examples/llm_inference.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -677,6 +681,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference_async.html b/latest/examples/llm_inference_async.html index 33ce47b8d0..e03351a886 100644 --- a/latest/examples/llm_inference_async.html +++ b/latest/examples/llm_inference_async.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -680,6 +684,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference_async_streaming.html b/latest/examples/llm_inference_async_streaming.html index 32559c9259..26a6874d88 100644 --- a/latest/examples/llm_inference_async_streaming.html +++ b/latest/examples/llm_inference_async_streaming.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -700,6 +704,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference_customize.html b/latest/examples/llm_inference_customize.html index 0dfda09e0e..ceb4b1aa83 100644 --- a/latest/examples/llm_inference_customize.html +++ b/latest/examples/llm_inference_customize.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -693,6 +697,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference_distributed.html b/latest/examples/llm_inference_distributed.html index 0c5554b237..2a9eaf5915 100644 --- a/latest/examples/llm_inference_distributed.html +++ b/latest/examples/llm_inference_distributed.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -681,6 +685,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_inference_kv_events.html b/latest/examples/llm_inference_kv_events.html index 5dd35528a3..d0de68e3b7 100644 --- a/latest/examples/llm_inference_kv_events.html +++ b/latest/examples/llm_inference_kv_events.html @@ -51,7 +51,7 @@ @@ -59,11 +59,11 @@ - + - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -511,53 +515,51 @@ 
     1### Get KV Cache Events
  2
  3from tensorrt_llm import LLM, SamplingParams
- 4from tensorrt_llm._torch.pyexecutor.config import PyTorchConfig
- 5from tensorrt_llm.llmapi import KvCacheConfig
+ 4from tensorrt_llm.llmapi import KvCacheConfig
+ 5
  6
- 7
- 8def main():
- 9    pytorch_config = PyTorchConfig(autotuner_enabled=False,
-10                                   kv_cache_dtype='auto')
-11
-12    llm = LLM(model="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
-13              tensor_parallel_size=2,
-14              pytorch_backend_config=pytorch_config,
-15              kv_cache_config=KvCacheConfig(enable_block_reuse=True,
-16                                            event_buffer_max_size=1024),
-17              backend="pytorch")
-18
-19    # Sample prompts having a common prefix.
-20    common_prefix = (
-21        "After the ghost's departure, Barnardo notes Horatio's pale appearance and asks if he's okay. "
-22        "Horatio concedes that he's shaken and confesses that, without witnessing the ghost himself, he wouldn't have believed it existed. "
-23        "He's also disturbed by the ghost's striking resemblance to the king. It even seems to be wearing the former king's armor. "
-24        "Horatio thinks the ghost's presence foretells that something is about to go wrong in Denmark. "
-25        "Marcellus concurs with Horatio, as he and the other guards have observed that their schedules have become more rigorous and have also noticed the preparations taking place within Elsinore, including the building of cannons, the storing of weapons, and the preparation of ships."
-26    )
-27    prompts = [
-28        common_prefix, common_prefix + " Marcellus also notes that the king's"
-29    ]
-30
-31    # Create a sampling params.
-32    sampling_params = SamplingParams(temperature=0.001,
-33                                     top_p=0.001,
-34                                     max_tokens=5)
-35
-36    for output in llm.generate(prompts, sampling_params=sampling_params):
-37        print(
-38            f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}"
-39        )
-40
-41    kv_events = llm.get_kv_cache_events(10)
-42    print(kv_events)
-43
-44    # Got output like follows:
-45    # [{'event_id': 0, 'data': {'type': 'created', 'num_blocks_per_cache_level': [101230, 0]}},
-46    #  {'event_id': 1, 'data': {'type': 'stored', 'parent_hash': None, 'blocks': [{'type': 'stored_block', 'block_hash': 4203099703668305365, 'tokens': [{'type': 'unique_token', 'token_id': 1, 'token_extra_id': 0}, ...
-47
-48
-49if __name__ == '__main__':
-50    main()
+ 7def main():
+ 8
+ 9    llm = LLM(model="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
+10              tensor_parallel_size=2,
+11              autotuner_enabled=False,
+12              kv_cache_dtype='auto',
+13              kv_cache_config=KvCacheConfig(enable_block_reuse=True,
+14                                            event_buffer_max_size=1024),
+15              backend="pytorch")
+16
+17    # Sample prompts having a common prefix.
+18    common_prefix = (
+19        "After the ghost's departure, Barnardo notes Horatio's pale appearance and asks if he's okay. "
+20        "Horatio concedes that he's shaken and confesses that, without witnessing the ghost himself, he wouldn't have believed it existed. "
+21        "He's also disturbed by the ghost's striking resemblance to the king. It even seems to be wearing the former king's armor. "
+22        "Horatio thinks the ghost's presence foretells that something is about to go wrong in Denmark. "
+23        "Marcellus concurs with Horatio, as he and the other guards have observed that their schedules have become more rigorous and have also noticed the preparations taking place within Elsinore, including the building of cannons, the storing of weapons, and the preparation of ships."
+24    )
+25    prompts = [
+26        common_prefix, common_prefix + " Marcellus also notes that the king's"
+27    ]
+28
+29    # Create a sampling params.
+30    sampling_params = SamplingParams(temperature=0.001,
+31                                     top_p=0.001,
+32                                     max_tokens=5)
+33
+34    for output in llm.generate(prompts, sampling_params=sampling_params):
+35        print(
+36            f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}"
+37        )
+38
+39    kv_events = llm.get_kv_cache_events(10)
+40    print(kv_events)
+41
+42    # Got output like follows:
+43    # [{'event_id': 0, 'data': {'type': 'created', 'num_blocks_per_cache_level': [101230, 0]}},
+44    #  {'event_id': 1, 'data': {'type': 'stored', 'parent_hash': None, 'blocks': [{'type': 'stored_block', 'block_hash': 4203099703668305365, 'tokens': [{'type': 'unique_token', 'token_id': 1, 'token_extra_id': 0}, ...
+45
+46
+47if __name__ == '__main__':
+48    main()
    @@ -573,12 +575,12 @@

    previous

    -

    Control generated text using logits processor

    +

    Generate Text Using Eagle2 Decoding
    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_logits_processor.html b/latest/examples/llm_logits_processor.html index 4a79379a87..a7a657defa 100644 --- a/latest/examples/llm_logits_processor.html +++ b/latest/examples/llm_logits_processor.html @@ -51,19 +51,19 @@ - + - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -525,106 +529,107 @@ 15# This simple callback will output a specific token at each step irrespective of prompt. 16# Refer to ../bindings/executor/example_logits_processor.py for a more 17# sophisticated callback that generates JSON structured output. - 18class MyLogitsProcessor(LogitsProcessor): - 19 - 20 def __init__(self, allowed_token_id: int): - 21 self.allowed_token_id = allowed_token_id - 22 - 23 def __call__(self, req_id: int, logits: torch.Tensor, - 24 token_ids: List[List[int]], stream_ptr: int, - 25 client_id: Optional[int]): - 26 mask = torch.full_like(logits, fill_value=float("-inf"), device="cpu") - 27 mask[:, :, self.allowed_token_id] = 0 - 28 - 29 stream = None if stream_ptr is None else torch.cuda.ExternalStream( - 30 stream_ptr) - 31 with torch.cuda.stream(stream): - 32 mask = mask.to(logits.device, non_blocking=True) - 33 logits += mask - 34 + 18# Please also refer to sampling_params.py for adding subclass to the approved class list for deserialization + 19class MyLogitsProcessor(LogitsProcessor): + 20 + 21 def __init__(self, allowed_token_id: int): + 22 self.allowed_token_id = allowed_token_id + 23 + 24 def __call__(self, req_id: int, logits: torch.Tensor, + 25 token_ids: List[List[int]], stream_ptr: int, + 26 client_id: Optional[int]): + 27 mask = torch.full_like(logits, fill_value=float("-inf"), device="cpu") + 28 mask[:, :, self.allowed_token_id] = 0 + 29 + 30 stream = None if stream_ptr is None else torch.cuda.ExternalStream( + 31 stream_ptr) + 32 with torch.cuda.stream(stream): + 33 mask = mask.to(logits.device, non_blocking=True) + 34 logits += mask 35 - 36# The recommended way to create a customized batched logits processor: - 37# * Subclass BatchedLogitsProcessor and implement the processing logics in the __call__ method. - 38# * Create an instance and pass to LLM. - 39# Alternatively, you can create any callable with the same signature with the __call__ method. - 40# A batched logits processor's arguments for all requests in a batch are made available as lists. - 41# This helps user optimize the callback for large batch sizes. For example: - 42# 1. Process more work on host, e.g. running a JSON state machine, in parallel with model forward pass on device. - 43# 2. Coalesce H2D memory transfers for all requests into a single cudaMemcpyAsync call. - 44# 3. Launch a single batched kernel, e.g. for updating logits on device. - 45class MyBatchedLogitsProcessor(BatchedLogitsProcessor): - 46 - 47 def __init__(self, allowed_token_id: int): - 48 self.allowed_token_id = allowed_token_id - 49 - 50 def __call__(self, req_ids: List[int], logits: List[torch.Tensor], - 51 token_ids: List[List[List[int]]], stream_ptr: int, - 52 client_ids: List[Optional[int]]): - 53 # Generate masks for all requests on host - 54 masks = [] - 55 for req_id, req_logits, req_token_ids, client_id in zip( - 56 req_ids, logits, token_ids, client_ids): - 57 mask = torch.full_like(req_logits, - 58 fill_value=float("-inf"), - 59 device="cpu") - 60 mask[:, :, self.allowed_token_id] = 0 - 61 masks.append(mask) - 62 - 63 # Move masks to device and add to logits using non-blocking operations - 64 with torch.cuda.stream(torch.cuda.ExternalStream(stream_ptr)): - 65 for req_logits, mask in zip(logits, masks): - 66 req_logits += mask.to(req_logits.device, non_blocking=True) - 67 + 36 + 37# The recommended way to create a customized batched logits processor: + 38# * Subclass BatchedLogitsProcessor and implement the processing logics in the __call__ method. + 39# * Create an instance and pass to LLM. + 40# Alternatively, you can create any callable with the same signature with the __call__ method. + 41# A batched logits processor's arguments for all requests in a batch are made available as lists. + 42# This helps user optimize the callback for large batch sizes. For example: + 43# 1. Process more work on host, e.g. running a JSON state machine, in parallel with model forward pass on device. + 44# 2. Coalesce H2D memory transfers for all requests into a single cudaMemcpyAsync call. + 45# 3. Launch a single batched kernel, e.g. for updating logits on device. + 46class MyBatchedLogitsProcessor(BatchedLogitsProcessor): + 47 + 48 def __init__(self, allowed_token_id: int): + 49 self.allowed_token_id = allowed_token_id + 50 + 51 def __call__(self, req_ids: List[int], logits: List[torch.Tensor], + 52 token_ids: List[List[List[int]]], stream_ptr: int, + 53 client_ids: List[Optional[int]]): + 54 # Generate masks for all requests on host + 55 masks = [] + 56 for req_id, req_logits, req_token_ids, client_id in zip( + 57 req_ids, logits, token_ids, client_ids): + 58 mask = torch.full_like(req_logits, + 59 fill_value=float("-inf"), + 60 device="cpu") + 61 mask[:, :, self.allowed_token_id] = 0 + 62 masks.append(mask) + 63 + 64 # Move masks to device and add to logits using non-blocking operations + 65 with torch.cuda.stream(torch.cuda.ExternalStream(stream_ptr)): + 66 for req_logits, mask in zip(logits, masks): + 67 req_logits += mask.to(req_logits.device, non_blocking=True) 68 - 69def main(): - 70 - 71 # Batched logits processor (only supported in TensorRT backend) - 72 # should be specified when initializing LLM. - 73 llm = LLM( - 74 model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", - 75 batched_logits_processor=MyBatchedLogitsProcessor(allowed_token_id=42)) - 76 - 77 # Sample prompts - 78 prompts = [ - 79 "Hello, my name is", - 80 "The president of the United States is", - 81 ] - 82 - 83 # Generate text - 84 for prompt_id, prompt in enumerate(prompts): - 85 # Use non-batched logits processor callback only for odd-numbered prompts - 86 if prompt_id % 2 == 0: - 87 sampling_params = SamplingParams(temperature=0.8, top_p=0.95) - 88 else: - 89 # Each prompt can be specified with a logits processor at runtime - 90 sampling_params = SamplingParams( - 91 temperature=0.8, - 92 top_p=0.95, - 93 logits_processor=MyLogitsProcessor(allowed_token_id=42)) - 94 - 95 for output in llm.generate([prompt], sampling_params): - 96 print( - 97 f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}" - 98 ) - 99 -100 # Got output like -101 # Prompt: 'Hello, my name is', Generated text: '\n\nJane Smith. I am a student pursuing my degree in Computer Science at [university]. I enjoy learning new things, especially technology and programming' -102 # Prompt: 'The president of the United States is', Generated text: "''''''''''''''''''''''''''''''''" -103 -104 # Use batched processor with batch size = 2 -105 sampling_params = SamplingParams(apply_batched_logits_processor=True) -106 for output in llm.generate(prompts, sampling_params): -107 print( -108 f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}" -109 ) -110 -111 # Got output like -112 # Prompt: 'Hello, my name is', Generated text: "''''''''''''''''''''''''''''''''" -113 # Prompt: 'The president of the United States is', Generated text: "''''''''''''''''''''''''''''''''" -114 + 69 + 70def main(): + 71 + 72 # Batched logits processor (only supported in TensorRT backend) + 73 # should be specified when initializing LLM. + 74 llm = LLM( + 75 model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", + 76 batched_logits_processor=MyBatchedLogitsProcessor(allowed_token_id=42)) + 77 + 78 # Sample prompts + 79 prompts = [ + 80 "Hello, my name is", + 81 "The president of the United States is", + 82 ] + 83 + 84 # Generate text + 85 for prompt_id, prompt in enumerate(prompts): + 86 # Use non-batched logits processor callback only for odd-numbered prompts + 87 if prompt_id % 2 == 0: + 88 sampling_params = SamplingParams(temperature=0.8, top_p=0.95) + 89 else: + 90 # Each prompt can be specified with a logits processor at runtime + 91 sampling_params = SamplingParams( + 92 temperature=0.8, + 93 top_p=0.95, + 94 logits_processor=MyLogitsProcessor(allowed_token_id=42)) + 95 + 96 for output in llm.generate([prompt], sampling_params): + 97 print( + 98 f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}" + 99 ) +100 +101 # Got output like +102 # Prompt: 'Hello, my name is', Generated text: '\n\nJane Smith. I am a student pursuing my degree in Computer Science at [university]. I enjoy learning new things, especially technology and programming' +103 # Prompt: 'The president of the United States is', Generated text: "''''''''''''''''''''''''''''''''" +104 +105 # Use batched processor with batch size = 2 +106 sampling_params = SamplingParams(apply_batched_logits_processor=True) +107 for output in llm.generate(prompts, sampling_params): +108 print( +109 f"Prompt: {output.prompt!r}, Generated text: {output.outputs[0].text!r}" +110 ) +111 +112 # Got output like +113 # Prompt: 'Hello, my name is', Generated text: "''''''''''''''''''''''''''''''''" +114 # Prompt: 'The president of the United States is', Generated text: "''''''''''''''''''''''''''''''''" 115 -116if __name__ == '__main__': -117 main() +116 +117if __name__ == '__main__': +118 main() @@ -649,11 +654,11 @@

    next

    -

    Get KV Cache Events

    +

    Generate Text Using Eagle2 Decoding

     @@ -754,6 +759,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_lookahead_decoding.html b/latest/examples/llm_lookahead_decoding.html index 296fb15b35..50c27b1e81 100644 --- a/latest/examples/llm_lookahead_decoding.html +++ b/latest/examples/llm_lookahead_decoding.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -675,6 +679,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_medusa_decoding.html b/latest/examples/llm_medusa_decoding.html index b38f8a6513..72f63e2800 100644 --- a/latest/examples/llm_medusa_decoding.html +++ b/latest/examples/llm_medusa_decoding.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -731,6 +735,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_mgmn_llm_distributed.html b/latest/examples/llm_mgmn_llm_distributed.html index d8f464d971..ca1735624e 100644 --- a/latest/examples/llm_mgmn_llm_distributed.html +++ b/latest/examples/llm_mgmn_llm_distributed.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -690,6 +694,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_mgmn_trtllm_bench.html b/latest/examples/llm_mgmn_trtllm_bench.html index 0a3c070501..c6f424b678 100644 --- a/latest/examples/llm_mgmn_trtllm_bench.html +++ b/latest/examples/llm_mgmn_trtllm_bench.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -584,25 +588,24 @@ 74 75 # This is optional 76 cat > /tmp/pytorch_extra_args.txt << EOF -77pytorch_backend_config: -78 use_cuda_graph: false -79 cuda_graph_padding_enabled: false -80 print_iter_log: true -81enable_attention_dp: false -82EOF -83 -84 # launch the benchmark -85 trtllm-llmapi-launch \ -86 trtllm-bench \ -87 --model $MODEL_NAME \ -88 --model_path $LOCAL_MODEL \ -89 throughput \ -90 --dataset $data_path \ -91 --backend pytorch \ -92 --tp 16 \ -93 --extra_llm_api_options /tmp/pytorch_extra_args.txt \ -94 $EXTRA_ARGS -95 " +77use_cuda_graph: false +78cuda_graph_padding_enabled: false +79print_iter_log: true +80enable_attention_dp: false +81EOF +82 +83 # launch the benchmark +84 trtllm-llmapi-launch \ +85 trtllm-bench \ +86 --model $MODEL_NAME \ +87 --model_path $LOCAL_MODEL \ +88 throughput \ +89 --dataset $data_path \ +90 --backend pytorch \ +91 --tp 16 \ +92 --extra_llm_api_options /tmp/pytorch_extra_args.txt \ +93 $EXTRA_ARGS +94 " @@ -732,6 +735,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_mgmn_trtllm_serve.html b/latest/examples/llm_mgmn_trtllm_serve.html index 6699772866..b5915e1fad 100644 --- a/latest/examples/llm_mgmn_trtllm_serve.html +++ b/latest/examples/llm_mgmn_trtllm_serve.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -692,6 +696,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_multilora.html b/latest/examples/llm_multilora.html index f667432cdd..0759100ce6 100644 --- a/latest/examples/llm_multilora.html +++ b/latest/examples/llm_multilora.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -696,6 +700,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/llm_quantization.html b/latest/examples/llm_quantization.html index 0b8f09ac85..ead01073f4 100644 --- a/latest/examples/llm_quantization.html +++ b/latest/examples/llm_quantization.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -718,6 +722,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/openai_chat_client.html b/latest/examples/openai_chat_client.html index 322a6e551c..24cd0bedbb 100644 --- a/latest/examples/openai_chat_client.html +++ b/latest/examples/openai_chat_client.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -659,6 +663,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/openai_chat_client_for_multimodal.html b/latest/examples/openai_chat_client_for_multimodal.html index 11824e4584..4d8104a432 100644 --- a/latest/examples/openai_chat_client_for_multimodal.html +++ b/latest/examples/openai_chat_client_for_multimodal.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -752,6 +756,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/openai_completion_client.html b/latest/examples/openai_completion_client.html index b8ed793432..8906617c30 100644 --- a/latest/examples/openai_completion_client.html +++ b/latest/examples/openai_completion_client.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -653,6 +657,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/examples/trtllm_serve_examples.html b/latest/examples/trtllm_serve_examples.html index 032e14d90a..e25b6bc98f 100644 --- a/latest/examples/trtllm_serve_examples.html +++ b/latest/examples/trtllm_serve_examples.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -353,6 +354,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -417,6 +419,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -451,6 +454,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -645,6 +649,15 @@

    +
    +
    + +

    Last updated on June 03, 2025.

    + +

    This page is generated by TensorRT-LLM commit 9ae2ce6.

    + +
    + diff --git a/latest/genindex.html b/latest/genindex.html index baa30c67e3..fb5b2ecb5d 100644 --- a/latest/genindex.html +++ b/latest/genindex.html @@ -50,7 +50,7 @@ @@ -60,7 +60,7 @@ - + @@ -329,6 +329,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -350,6 +351,7 @@
  • Generate Text Asynchronously
  • Distributed LLM Generation
  • Control generated text using logits processor
    • +
  • Generate Text Using Eagle2 Decoding
  • Get KV Cache Events
  • Generate Text Using Lookahead Decoding
  • Generation with Quantization
    • @@ -414,6 +416,7 @@
  • Graph Rewriting Module
  • Run gpt-2b + LoRA using Executor / cpp runtime
  • Expert Parallelism in TensorRT-LLM
    • +
  • KV Cache Management: Pools, Blocks, and Events
  • KV cache reuse
  • Speculative Sampling
  • Disaggregated-Service (experimental)
    • @@ -448,6 +451,7 @@
  • Speed up inference with SOTA quantization techniques in TRT-LLM
  • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
  • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
    • +
  • DeepSeek R1 MTP Implementation and Optimization
    • @@ -775,14 +779,14 @@
  • AllReduceParams (class in tensorrt_llm.functional)
    • - - + - +
  • conv_transpose2d() (in module tensorrt_llm.functional) +
    • +
  • convert_load_format() (tensorrt_llm.llmapi.TorchLlmArgs class method)
  • ConvTranspose2d (class in tensorrt_llm.layers.conv)
    • @@ -1112,8 +1140,12 @@
  • cross_kv_cache_fraction (tensorrt_llm.llmapi.KvCacheConfig attribute)
  • ctx_request_id (tensorrt_llm.llmapi.DisaggregatedParams attribute) +
    • +
  • cuda_graph_batch_sizes (tensorrt_llm.llmapi.TorchLlmArgs attribute)
  • cuda_graph_cache_size (tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig attribute) +
    • +
  • cuda_graph_max_batch_size (tensorrt_llm.llmapi.TorchLlmArgs attribute)
  • cuda_graph_mode (tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig attribute) @@ -1121,6 +1153,8 @@
  • (tensorrt_llm.runtime.GenerationSession attribute)
    • +
  • cuda_graph_padding_enabled (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
  • cuda_stream_guard() (tensorrt_llm.runtime.GenerationSession method)
  • cuda_stream_sync() (in module tensorrt_llm.functional) @@ -1163,6 +1197,12 @@
  • DecoderModel (class in tensorrt_llm.models)
    • +
  • decoding_config (tensorrt_llm.llmapi.TorchLlmArgs attribute) + +
  • decoding_type (tensorrt_llm.llmapi.EagleDecodingConfig attribute)
  • DeepseekForCausalLM (class in tensorrt_llm.models) @@ -1200,10 +1242,14 @@
  • DiffusersAttention (class in tensorrt_llm.layers.attention)
  • DimRange (class in tensorrt_llm.functional) +
    • +
  • directory (tensorrt_llm.llmapi.KvCacheRetentionConfig property)
  • disable (tensorrt_llm.functional.SideStreamIDType attribute)
  • disable_forward_chunking() (tensorrt_llm.models.SD3Transformer2DModel method) +
    • +
  • disable_overlap_scheduler (tensorrt_llm.llmapi.TorchLlmArgs attribute)
  • disaggregated_params (tensorrt_llm.llmapi.CompletionOutput attribute)
    • @@ -1256,6 +1302,8 @@
    @@ -1348,6 +1416,8 @@
  • FalconForCausalLM (class in tensorrt_llm.models)
  • FalconModel (class in tensorrt_llm.models) +
    • +
  • fast_build (tensorrt_llm.llmapi.TrtLlmArgs attribute)
  • fc_gate() (tensorrt_llm.layers.mlp.FusedGatedMLP method)
    • @@ -1357,6 +1427,12 @@
  • fc_gate_plugin() (tensorrt_llm.layers.mlp.FusedGatedMLP method)
    • +
  • field_name (tensorrt_llm.llmapi.TorchLlmArgs attribute), [1], [2], [3] + +
  • fill_attention_const_params_for_long_rope() (tensorrt_llm.layers.attention.AttentionParams method)
  • fill_attention_const_params_for_rope() (tensorrt_llm.layers.attention.AttentionParams method) @@ -1571,6 +1647,8 @@
  • (tensorrt_llm.llmapi.MedusaDecodingConfig class method)
  • (tensorrt_llm.llmapi.MTPDecodingConfig class method) +
    • +
  • (tensorrt_llm.llmapi.NGramDecodingConfig class method)
  • (tensorrt_llm.llmapi.QuantConfig class method)
    • @@ -1814,6 +1892,8 @@
  • get_num_heads_kv() (tensorrt_llm.runtime.GenerationSession method)
  • get_parent() (tensorrt_llm.functional.Tensor method) +
    • +
  • get_pytorch_backend_config() (tensorrt_llm.llmapi.TorchLlmArgs method)
  • get_request_type() (tensorrt_llm.llmapi.DisaggregatedParams method)
    • @@ -1963,12 +2043,12 @@
  • int_clip() (in module tensorrt_llm.functional)
    • - - +
    • is_deferred() (tensorrt_llm.functional.PositionEmbeddingType method)
    • is_dynamic() (tensorrt_llm.functional.Tensor method) @@ -1978,18 +2058,24 @@
    • is_gemma_2 (tensorrt_llm.models.GemmaConfig property)
    • is_gemma_3 (tensorrt_llm.models.GemmaConfig property) +
      • +
    • is_keep_all (tensorrt_llm.llmapi.NGramDecodingConfig attribute)
    • is_medusa_mode (tensorrt_llm.runtime.GenerationSession property)
    • is_module_excluded_from_quantization() (tensorrt_llm.llmapi.QuantConfig method)
    • is_mrope() (tensorrt_llm.functional.PositionEmbeddingType method) +
      • +
    • is_public_pool (tensorrt_llm.llmapi.NGramDecodingConfig attribute)
    • is_redrafter_mode (tensorrt_llm.runtime.GenerationSession property)
    • is_rope() (tensorrt_llm.functional.PositionEmbeddingType method)
    • is_trt_wrapper() (tensorrt_llm.functional.Tensor method) +
      • +
    • is_use_oldest (tensorrt_llm.llmapi.NGramDecodingConfig attribute)
    • is_valid() (tensorrt_llm.layers.attention.AttentionParams method) @@ -2020,6 +2106,8 @@ - + - + + -
      +
    • load_format (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
    • load_test_audio() (tensorrt_llm.runtime.MultimodalModelRunner method)
    • load_test_data() (tensorrt_llm.runtime.MultimodalModelRunner method) @@ -2182,6 +2274,8 @@
    • low_latency_gemm() (in module tensorrt_llm.functional)
    • low_latency_gemm_swiglu() (in module tensorrt_llm.functional) +
      • +
    • LOWPRECISION (tensorrt_llm.functional.AllReduceStrategy attribute)
    • lt() (in module tensorrt_llm.functional)
      • @@ -2239,6 +2333,12 @@ +
    • max_cpu_loras (tensorrt_llm.llmapi.TorchLlmArgs attribute) + +
    • max_draft_len (tensorrt_llm.llmapi.BuildConfig attribute) @@ -2248,6 +2348,20 @@
    • max_encoder_input_len (tensorrt_llm.llmapi.BuildConfig attribute)
    • max_input_len (tensorrt_llm.llmapi.BuildConfig attribute) +
      • +
    • max_lora_rank (tensorrt_llm.llmapi.TorchLlmArgs attribute) + +
      • +
    • max_loras (tensorrt_llm.llmapi.TorchLlmArgs attribute) + +
      • +
    • max_matching_ngram_size (tensorrt_llm.llmapi.NGramDecodingConfig attribute)
    • max_medusa_tokens (tensorrt_llm.runtime.ModelConfig attribute)
      • @@ -2335,12 +2449,12 @@
    • min() (in module tensorrt_llm.functional)
      • -
    • MropeParams (class in tensorrt_llm.layers.attention)
      • +
    • msg (tensorrt_llm.llmapi.TorchLlmArgs attribute), [1], [2], [3] + +
    • MTPDecodingConfig (class in tensorrt_llm.llmapi)
    • mul() (in module tensorrt_llm.functional) @@ -2498,6 +2638,10 @@
    • network (tensorrt_llm.functional.Tensor property)
    • next_medusa_input_ids() (tensorrt_llm.runtime.GenerationSession method) +
      • +
    • NGRAM (tensorrt_llm.models.SpeculativeDecodingMode attribute) +
      • +
    • NGramDecodingConfig (class in tensorrt_llm.llmapi)
    • NO_QUANT (tensorrt_llm.llmapi.QuantAlgo attribute)
      • @@ -2520,11 +2664,11 @@
    • not_op() (in module tensorrt_llm.functional)
    • num_beams (tensorrt_llm.runtime.SamplingConfig attribute) -
      • -
    • num_draft_tokens (tensorrt_llm.runtime.GenerationSession attribute)
      		• +
    • prompt_lookup_num_tokens (tensorrt_llm.llmapi.NGramDecodingConfig attribute) +
    • prompt_token_ids (tensorrt_llm.llmapi.RequestOutput attribute)
    • PromptTuningEmbedding (class in tensorrt_llm.layers.embedding) @@ -3228,7 +3376,7 @@
    • module, [1], [2], [3], [4], [5]
      • -
    • tensorrt_llm (C++ type), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41] +
    • tensorrt_llm (C++ type), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42]
    • tensorrt_llm.functional @@ -3328,11 +3476,11 @@
    • module
      • -
    • tensorrt_llm::batch_manager (C++ type), [1], [2], [3], [4], [5] +
    • tensorrt_llm::batch_manager (C++ type), [1], [2], [3], [4]
    • tensorrt_llm::batch_manager::kv_cache_manager (C++ type)
      • -
    • tensorrt_llm::executor (C++ type), [1], [2], [3], [4], [5], [6] +
    • tensorrt_llm::executor (C++ type), [1], [2], [3], [4], [5], [6], [7]
    • tensorrt_llm::executor::AdditionalModelOutput (C++ class)
      • @@ -4239,6 +4387,8 @@
    • tensorrt_llm::executor::IterationStats::numQueuedRequests (C++ member)
    • tensorrt_llm::executor::IterationStats::pinnedMemUsage (C++ member) +
      • +
    • tensorrt_llm::executor::IterationStats::specDecStats (C++ member)
    • tensorrt_llm::executor::IterationStats::staticBatchingStats (C++ member)
      • @@ -4250,7 +4400,59 @@
    • tensorrt_llm::executor::JsonSerialization::toJsonStr (C++ function), [1], [2]
      • -
    • tensorrt_llm::executor::kv_cache (C++ type), [1], [2] +
    • tensorrt_llm::executor::kv_cache (C++ type), [1], [2], [3] +
      • +
    • tensorrt_llm::executor::kv_cache::AgentDesc (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentDesc::AgentDesc (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentDesc::getBackendAgentDesc (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentDesc::mBackendAgentDesc (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState (C++ struct) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState::AgentState (C++ function), [1] +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState::mAgentName (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState::mConnectionInfo (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState::operator== (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::AgentState::toString (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseAgentConfig (C++ struct) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseAgentConfig::mName (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseAgentConfig::useProgThread (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::checkRemoteDescs (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::connectRemoteAgent (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::deregisterMemory (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::getConnectionInfo (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::getLocalAgentDesc (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::getNotifiedSyncMessages (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::invalidateRemoteAgent (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::loadRemoteAgent (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::notifySyncMessage (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::registerMemory (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::submitTransferRequests (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::BaseTransferAgent::~BaseTransferAgent (C++ function)
    • tensorrt_llm::executor::kv_cache::CacheState (C++ class)
      • @@ -4316,13 +4518,17 @@
    • tensorrt_llm::executor::kv_cache::CommState (C++ class)
      • -
    • tensorrt_llm::executor::kv_cache::CommState::CommState (C++ function), [1], [2], [3] +
    • tensorrt_llm::executor::kv_cache::CommState::CommState (C++ function), [1], [2], [3], [4] +
      • +
    • tensorrt_llm::executor::kv_cache::CommState::getAgentState (C++ function)
    • tensorrt_llm::executor::kv_cache::CommState::getMpiState (C++ function)
    • tensorrt_llm::executor::kv_cache::CommState::getSelfIdx (C++ function)
    • tensorrt_llm::executor::kv_cache::CommState::getSocketState (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::CommState::isAgentState (C++ function)
    • tensorrt_llm::executor::kv_cache::CommState::isMpiState (C++ function)
      • @@ -4345,6 +4551,8 @@
    • tensorrt_llm::executor::kv_cache::Connection::send (C++ function)
    • tensorrt_llm::executor::kv_cache::Connection::~Connection (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::ConnectionInfoType (C++ type)
    • tensorrt_llm::executor::kv_cache::ConnectionManager (C++ class)
      • @@ -4363,6 +4571,74 @@
    • tensorrt_llm::executor::kv_cache::DataContext::getTag (C++ function)
    • tensorrt_llm::executor::kv_cache::DataContext::mTag (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::dlSym (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::DynLibLoader (C++ function), [1] +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::getFunctionPointer (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::getHandle (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::getInstance (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::mDllMutex (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::mHandlers (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::operator= (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::DynLibLoader::~DynLibLoader (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::makeTransferAgent (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::deserialize (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::getAddr (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::getDeviceId (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::getLen (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::mAddr (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::mDeviceId (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc (C++ function), [1], [2] +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::mLen (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::serialize (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDesc::serializedSize (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs::getDescs (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs::getType (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs::mDescs (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs::MemoryDescs (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryDescs::mType (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType (C++ enum) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType::kBLK (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType::kDRAM (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType::kFILE (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType::kOBJ (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::MemoryType::kVRAM (C++ enumerator)
    • tensorrt_llm::executor::kv_cache::MpiState (C++ struct)
      • @@ -4371,6 +4647,8 @@
    • tensorrt_llm::executor::kv_cache::MpiState::operator== (C++ function)
    • tensorrt_llm::executor::kv_cache::MpiState::toString (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::RegisterDescs (C++ type)
    • tensorrt_llm::executor::kv_cache::SocketState (C++ struct)
      • @@ -4381,6 +4659,48 @@
    • tensorrt_llm::executor::kv_cache::SocketState::operator== (C++ function)
    • tensorrt_llm::executor::kv_cache::SocketState::toString (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::SyncMessage (C++ type) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferDescs (C++ type) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferOp (C++ enum) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferOp::kREAD (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferOp::kWRITE (C++ enumerator) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::getDstDescs (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::getOp (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::getRemoteName (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::getSrcDescs (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::getSyncMessage (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::mDstDescs (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::mOp (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::mRemoteName (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::mSrcDescs (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::mSyncMessage (C++ member) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferStatus (C++ class) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferStatus::isCompleted (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferStatus::wait (C++ function) +
      • +
    • tensorrt_llm::executor::kv_cache::TransferStatus::~TransferStatus (C++ function)
    • tensorrt_llm::executor::KvCacheConfig (C++ class)
      • @@ -4497,10 +4817,14 @@
    • tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeDurationMs (C++ function)
    • tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeRetentionPriority (C++ function) +
      • +
    • tensorrt_llm::executor::KvCacheRetentionConfig::getDirectory (C++ function)
    • tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration (C++ function)
    • tensorrt_llm::executor::KvCacheRetentionConfig::getTokenRangeRetentionConfigs (C++ function) +
      • +
    • tensorrt_llm::executor::KvCacheRetentionConfig::getTransferMode (C++ function)
    • tensorrt_llm::executor::KvCacheRetentionConfig::kDefaultRetentionPriority (C++ member)
      • @@ -4508,13 +4832,17 @@
    • tensorrt_llm::executor::KvCacheRetentionConfig::kMinRetentionPriority (C++ member)
      • -
    • tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig (C++ function), [1] +
    • tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig (C++ function), [1]
    • tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeDurationMs (C++ member)
    • tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeRetentionPriority (C++ member) +
      • +
    • tensorrt_llm::executor::KvCacheRetentionConfig::mDirectory (C++ member)
    • tensorrt_llm::executor::KvCacheRetentionConfig::mTokenRangeRetentionConfigs (C++ member) +
      • +
    • tensorrt_llm::executor::KvCacheRetentionConfig::mTransferMode (C++ member)
    • tensorrt_llm::executor::KvCacheRetentionConfig::operator== (C++ function)
      • @@ -4571,6 +4899,14 @@
    • tensorrt_llm::executor::KVCacheStoredData::blocks (C++ member)
    • tensorrt_llm::executor::KVCacheStoredData::parentHash (C++ member) +
      • +
    • tensorrt_llm::executor::KvCacheTransferMode (C++ enum) +
      • +
    • tensorrt_llm::executor::KvCacheTransferMode::DRAM (C++ enumerator) +
      • +
    • tensorrt_llm::executor::KvCacheTransferMode::GDS (C++ enumerator) +
      • +
    • tensorrt_llm::executor::KvCacheTransferMode::POSIX_DEBUG_FALLBACK (C++ enumerator)
    • tensorrt_llm::executor::KVCacheUpdatedData (C++ struct)
      • @@ -5367,6 +5703,8 @@
    • tensorrt_llm::executor::Serialization::deserializeAdditionalModelOutput (C++ function)
    • tensorrt_llm::executor::Serialization::deserializeAdditionalOutput (C++ function) +
      • +
    • tensorrt_llm::executor::Serialization::deserializeAgentState (C++ function)
    • tensorrt_llm::executor::Serialization::deserializeBool (C++ function)
      • @@ -5457,6 +5795,8 @@
    • tensorrt_llm::executor::Serialization::deserializeSocketState (C++ function)
    • tensorrt_llm::executor::Serialization::deserializeSpecDecFastLogitsInfo (C++ function) +
      • +
    • tensorrt_llm::executor::Serialization::deserializeSpecDecodingStats (C++ function)
    • tensorrt_llm::executor::Serialization::deserializeSpeculativeDecodingConfig (C++ function)
      • @@ -5470,9 +5810,9 @@
    • tensorrt_llm::executor::Serialization::deserializeTokenRangeRetentionConfig (C++ function)
      • -
    • tensorrt_llm::executor::Serialization::serialize (C++ function), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52] +
    • tensorrt_llm::executor::Serialization::serialize (C++ function), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54]
      • -
    • tensorrt_llm::executor::Serialization::serializedSize (C++ function), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46] +
    • tensorrt_llm::executor::Serialization::serializedSize (C++ function), [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48]
    • tensorrt_llm::executor::Shape (C++ class)
      • @@ -5483,6 +5823,22 @@
    • tensorrt_llm::executor::Shape::Shape (C++ function), [1], [2]
    • tensorrt_llm::executor::SizeType32 (C++ type) +
      • +
    • tensorrt_llm::executor::SizeType64 (C++ type) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats (C++ struct) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::acceptanceLength (C++ member) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::draftOverhead (C++ member) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::iterLatencyMS (C++ member) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::numAcceptedTokens (C++ member) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::numDraftTokens (C++ member) +
      • +
    • tensorrt_llm::executor::SpecDecodingStats::numRequestsWithDraftTokens (C++ member)
    • tensorrt_llm::executor::SpeculativeDecodingConfig (C++ class)
      • @@ -5637,6 +5993,8 @@
    • tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers (C++ function)
    • tensorrt_llm::runtime::AllReduceBuffers::mAllReduceCommPtrs (C++ member) +
      • +
    • tensorrt_llm::runtime::AllReduceBuffers::mFlagPtrs (C++ member)
    • tensorrt_llm::runtime::AllReduceBuffers::mIpcMemoryHandles (C++ member)
      • @@ -5750,6 +6108,8 @@
    • tensorrt_llm::runtime::CudaEvent::Deleter::mOwnsEvent (C++ member)
      • +
    • topk() (in module tensorrt_llm.functional) +
      • +
    • torch_compile_enable_userbuffers (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
      • +
    • torch_compile_enabled (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
      • +
    • torch_compile_fullgraph (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
      • +
    • torch_compile_inductor_enabled (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
      • +
    • torch_compile_piecewise_cuda_graph (tensorrt_llm.llmapi.TorchLlmArgs attribute) +
      • +
    • TorchLlmArgs (class in tensorrt_llm.llmapi)
    • tp_split_dim() (tensorrt_llm.layers.linear.Linear class method) @@ -8386,6 +8760,8 @@
    • (tensorrt_llm.layers.linear.RowLinear class method)
      • +
    • transfer_mode (tensorrt_llm.llmapi.KvCacheRetentionConfig property) +
    • transpose() (in module tensorrt_llm.functional)
        @@ -8466,6 +8842,8 @@
    • trtllm_modules_to_hf_modules (tensorrt_llm.runtime.ModelConfig attribute) +
      • +
    • TrtLlmArgs (class in tensorrt_llm.llmapi)
    • truncate_prompt_tokens (tensorrt_llm.llmapi.SamplingParams attribute)
      • @@ -8514,6 +8892,8 @@
    • use_beam_hyps (tensorrt_llm.runtime.SamplingConfig attribute)
    • use_beam_search (tensorrt_llm.llmapi.SamplingParams attribute) +
      • +
    • use_cuda_graph (tensorrt_llm.llmapi.TorchLlmArgs attribute)
    • use_dynamic_tree (tensorrt_llm.llmapi.EagleDecodingConfig attribute)
      • @@ -8523,8 +8903,12 @@
      • use_gpt_attention_plugin (tensorrt_llm.runtime.GenerationSession property)
        • -
      • use_kv_cache (tensorrt_llm.runtime.GenerationSession property) +
      • use_kv_cache (tensorrt_llm.llmapi.TorchLlmArgs attribute) + +
      • use_lora() (tensorrt_llm.models.DecoderModel method)
          @@ -8569,6 +8953,10 @@

          V

          - + @@ -8784,6 +9180,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/index.html b/latest/index.html index c4e44cda0f..a38bd94cb7 100644 --- a/latest/index.html +++ b/latest/index.html @@ -51,7 +51,7 @@ @@ -62,7 +62,7 @@ - + @@ -335,6 +335,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -356,6 +357,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -420,6 +422,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -454,6 +457,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -641,6 +645,11 @@
        • How to Enable
          • +
        • KV Cache Management: Pools, Blocks, and Events +
        • KV cache reuse @@ -848,6 +852,15 @@ pip install ./build/tensorrt_llm*.

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/installation/grace-hopper.html b/latest/installation/grace-hopper.html index 6705d95cb5..a88006a446 100644 --- a/latest/installation/grace-hopper.html +++ b/latest/installation/grace-hopper.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -353,6 +354,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -417,6 +419,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -451,6 +454,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -672,6 +676,15 @@ sudo apt-get -y +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/installation/linux.html b/latest/installation/linux.html index 060cf3df86..a3beeadce7 100644 --- a/latest/installation/linux.html +++ b/latest/installation/linux.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -353,6 +354,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -417,6 +419,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -451,6 +454,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -708,6 +712,15 @@ Please install CUDA toolkit when you see the following message when running Mode

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/key-features.html b/latest/key-features.html index 0776eb158c..73ad3a2bcf 100644 --- a/latest/key-features.html +++ b/latest/key-features.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -332,6 +332,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -353,6 +354,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -417,6 +419,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -451,6 +454,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -640,6 +644,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/llm-api/index.html b/latest/llm-api/index.html index 69ff46e942..e89c88f65c 100644 --- a/latest/llm-api/index.html +++ b/latest/llm-api/index.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -755,6 +759,15 @@ Refer to the +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/llm-api/reference.html b/latest/llm-api/reference.html index 82ffe4010e..8ee33f8585 100644 --- a/latest/llm-api/reference.html +++ b/latest/llm-api/reference.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -548,41 +552,39 @@

        • moe_expert_parallel_size (Optional[int]) – The expert parallel size for MoE models’s expert weights. Defaults to None.

        • enable_attention_dp (bool) – Enable attention data parallel. Defaults to False.

        • cp_config (Optional[dict]) – Context parallel config. Defaults to None.

          • -

        • auto_parallel (bool) – Enable auto parallel mode. Defaults to False.

          • -

        • auto_parallel_world_size (Optional[int]) – The world size for auto parallel mode. Defaults to None.

        • load_format (Literal['auto', 'dummy']) – The format to load the model. Defaults to auto.

          • -

        • enable_tqdm (bool) – Enable tqdm for progress bar. Defaults to False.

        • enable_lora (bool) – Enable LoRA. Defaults to False.

        • lora_config (Optional[tensorrt_llm.lora_manager.LoraConfig]) – LoRA configuration for the model. Defaults to None.

        • enable_prompt_adapter (bool) – Enable prompt adapter. Defaults to False.

        • max_prompt_adapter_token (int) – The maximum number of prompt adapter tokens. Defaults to 0.

        • quant_config (Optional[tensorrt_llm.models.modeling_utils.QuantConfig]) – Quantization config. Defaults to None.

          • -

        • calib_config (Optional[tensorrt_llm.llmapi.llm_args.CalibConfig]) – Calibration config. Defaults to None.

          • -

        • build_config (Optional[tensorrt_llm.builder.BuildConfig]) – Build config. Defaults to None.

          • -

        • kv_cache_config (Optional[tensorrt_llm.llmapi.llm_args.KvCacheConfig]) – KV cache config. Defaults to None.

          • +

        • kv_cache_config (tensorrt_llm.llmapi.llm_args.KvCacheConfig) – KV cache config. Defaults to None.

        • enable_chunked_prefill (bool) – Enable chunked prefill. Defaults to False.

        • guided_decoding_backend (Optional[str]) – Guided decoding backend. Defaults to None.

        • batched_logits_processor (Optional[tensorrt_llm.sampling_params.BatchedLogitsProcessor]) – Batched logits processor. Defaults to None.

        • iter_stats_max_iterations (Optional[int]) – The maximum number of iterations for iter stats. Defaults to None.

        • request_stats_max_iterations (Optional[int]) – The maximum number of iterations for request stats. Defaults to None.

          • -

        • workspace (Optional[str]) – The workspace for the model. Defaults to None.

          • -

        • embedding_parallel_mode (str) – The embedding parallel mode. Defaults to SHARDING_ALONG_VOCAB.

          • -

        • fast_build (bool) – Enable fast build. Defaults to False.

          • -

        • enable_build_cache (Union[tensorrt_llm.llmapi.build_cache.BuildCacheConfig, bool]) – Enable build cache. Defaults to False.

        • peft_cache_config (Optional[tensorrt_llm.llmapi.llm_args.PeftCacheConfig]) – PEFT cache config. Defaults to None.

          • -

        • scheduler_config (Optional[tensorrt_llm.llmapi.llm_args.SchedulerConfig]) – Scheduler config. Defaults to None.

          • +

        • scheduler_config (tensorrt_llm.llmapi.llm_args.SchedulerConfig) – Scheduler config. Defaults to None.

        • cache_transceiver_config (Optional[tensorrt_llm.llmapi.llm_args.CacheTransceiverConfig]) – Cache transceiver config. Defaults to None.

          • -

        • speculative_config (Union[tensorrt_llm.llmapi.llm_args.LookaheadDecodingConfig, tensorrt_llm.llmapi.llm_args.MedusaDecodingConfig, tensorrt_llm.llmapi.llm_args.EagleDecodingConfig, tensorrt_llm.llmapi.llm_args.MTPDecodingConfig, NoneType]) – Speculative decoding config. Defaults to None.

          • +

        • speculative_config (Union[tensorrt_llm.llmapi.llm_args.LookaheadDecodingConfig, tensorrt_llm.llmapi.llm_args.MedusaDecodingConfig, tensorrt_llm.llmapi.llm_args.EagleDecodingConfig, tensorrt_llm.llmapi.llm_args.MTPDecodingConfig, tensorrt_llm.llmapi.llm_args.NGramDecodingConfig, NoneType]) – Speculative decoding config. Defaults to None.

        • batching_type (Optional[tensorrt_llm.llmapi.llm_args.BatchingType]) – Batching type. Defaults to None.

        • normalize_log_probs (bool) – Normalize log probabilities. Defaults to False.

          • -

        • gather_generation_logits (bool) – Gather generation logits. Defaults to False.

          • -

        • extended_runtime_perf_knob_config (Optional[tensorrt_llm.llmapi.llm_args.ExtendedRuntimePerfKnobConfig]) – Extended runtime perf knob config. Defaults to None.

        • max_batch_size (Optional[int]) – The maximum batch size. Defaults to None.

        • max_input_len (int) – The maximum input length. Defaults to 1024.

        • max_seq_len (Optional[int]) – The maximum sequence length. Defaults to None.

        • max_beam_width (int) – The maximum beam width. Defaults to 1.

        • max_num_tokens (Optional[int]) – The maximum number of tokens. Defaults to None.

        • backend (Optional[str]) – The backend to use. Defaults to None.

          • +

        • gather_generation_logits (bool) – Gather generation logits. Defaults to False.

          • +

        • enable_tqdm (bool) – Enable tqdm for progress bar. Defaults to False.

          • +

        • build_config (Optional[tensorrt_llm.builder.BuildConfig]) – Build config. Defaults to None.

          • +

        • workspace (Optional[str]) – The workspace for the model. Defaults to None.

          • +

        • enable_build_cache (Union[tensorrt_llm.llmapi.build_cache.BuildCacheConfig, bool]) – Enable build cache. Defaults to False.

          • +

        • extended_runtime_perf_knob_config (Optional[tensorrt_llm.llmapi.llm_args.ExtendedRuntimePerfKnobConfig]) – Extended runtime perf knob config. Defaults to None.

          • +

        • calib_config (Optional[tensorrt_llm.llmapi.llm_args.CalibConfig]) – Calibration config. Defaults to None.

          • +

        • embedding_parallel_mode (str) – The embedding parallel mode. Defaults to SHARDING_ALONG_VOCAB.

          • +

        • fast_build (bool) – Enable fast build. Defaults to False.

        • kwargs (Any) – Advanced arguments passed to LlmArgs.

          • @@ -1860,6 +1862,8 @@ The BatchedLogitsProcessor class is recommended for callback creation. The callb
          token_range_retention_configs: list[tensorrt_llm.bindings.executor.KvCacheRetentionConfig.TokenRangeRetentionConfig],
          decode_retention_priority: int = 35,
          decode_duration_ms: datetime.timedelta | None = None,
          +
          transfer_mode: tensorrt_llm.bindings.executor.KvCacheTransferMode = DRAM,
          +
          directory: str | None = None,
          ) → None# @@ -1875,11 +1879,21 @@ The BatchedLogitsProcessor class is recommended for callback creation. The callb property decode_retention_priority#
          +
          +
          +property directory#
          +
          +
          property token_range_retention_configs#
          +
          +
          +property transfer_mode#
          +
          +
          @@ -1926,18 +1940,39 @@ validated to form a valid model.

          field max_ngram_size: int = 3#

          Number of tokens per NGram.

          +
          +
          Validated by:
          +
            +

          • validate_positive_values

            • +
          +
          +
          field max_verification_set_size: int = 4#

          Number of NGrams in verification branch per step.

          +
          +
          Validated by:
          +
            +

          • validate_positive_values

            • +
          +
          +
          field max_window_size: int = 4#

          Number of NGrams in lookahead branch per step.

          +
          +
          Validated by:
          +
            +

          • validate_positive_values

            • +
          +
          +
          @@ -1946,9 +1981,9 @@ validated to form a valid model.

          Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

          -
          +
          -classmethod validate_positive_values(v)[source]#
          +validator validate_positive_values  »  max_ngram_size, max_window_size, max_verification_set_size[source]#
          @@ -2011,6 +2046,7 @@ validated to form a valid model.

          num_eagle_layers: int | None = None,
          max_non_leaves_per_layer: int | None = None,
          pytorch_eagle_weights_path: str | None = None,
          +
          eagle3_one_model: bool | None = True,
          )[source]# @@ -2025,6 +2061,11 @@ validated to form a valid model.

          field dynamic_tree_max_topK: int | None = None#
          +
          +
          +field eagle3_one_model: bool | None = True#
          +
          +
          field eagle_choices: List[List[int]] | None = None#
          @@ -3172,6 +3213,957 @@ changed, you should remove the caches manually.

          +
          +
          +class tensorrt_llm.llmapi.NGramDecodingConfig( + +
          +
          *,
          +
          max_draft_len: int | None = None,
          +
          speculative_model: str | Path | None = None,
          +
          prompt_lookup_num_tokens: int = 2,
          +
          max_matching_ngram_size: int = 4,
          +
          is_keep_all: bool = True,
          +
          is_use_oldest: bool = True,
          +
          is_public_pool: bool = True,
          +
          + +)[source]#
          +

          Bases: DecodingBaseConfig

          +

          Configuration for NGram drafter speculative decoding.

          +
          +
          Parameters:
          +
            +

          • prompt_lookup_num_tokens – int +The length maximum of draft tokens (can be understood as length maximum of output draft tokens).

            • +

          • max_matching_ngram_size – int +The length maximum of searching tokens (can be understood as length maximum of input tokens to search).

            • +

          • is_keep_all – bool = True +Whether to keep all candidate pattern-matches pairs, only one match is kept for each pattern if False.

            • +

          • is_use_oldest – bool = True +Whether to provide the oldest match when pattern is hit, the newest one is provided if False.

            • +

          • is_public_pool – bool = True +Whether to use a common pool for all requests, or the pool is private for each request if False.

            • +
          +
          +
          +
          +
          +decoding_type: ClassVar[str] = 'NGram'#
          +
          + +
          +
          +classmethod from_dict(data: dict)[source]#
          +
          + +
          +
          +field is_keep_all: bool = True#
          +
          + +
          +
          +field is_public_pool: bool = True#
          +
          + +
          +
          +field is_use_oldest: bool = True#
          +
          + +
          +
          +field max_matching_ngram_size: int = 4#
          +
          + +
          +
          +model_config: ClassVar[ConfigDict] = {}#
          +

          Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

          +
          + +
          +
          +field prompt_lookup_num_tokens: int = 2#
          +
          + +
          + +
          +
          +tensorrt_llm.llmapi.LlmArgs#
          +

          alias of TrtLlmArgs

          +
          + +
          +
          +class tensorrt_llm.llmapi.TorchLlmArgs( + +
          +
          *,
          +
          model: str | ~pathlib.Path,
          +
          tokenizer: str | ~pathlib.Path | ~transformers.tokenization_utils_base.PreTrainedTokenizerBase | ~tensorrt_llm.llmapi.tokenizer.TokenizerBase | None = None,
          +
          tokenizer_mode: ~typing.Literal['auto',
          +
          'slow'] = 'auto',
          +
          skip_tokenizer_init: bool = False,
          +
          trust_remote_code: bool = False,
          +
          tensor_parallel_size: int = 1,
          +
          dtype: str = 'auto',
          +
          revision: str | None = None,
          +
          tokenizer_revision: str | None = None,
          +
          pipeline_parallel_size: int = 1,
          +
          context_parallel_size: int = 1,
          +
          gpus_per_node: int | None = None,
          +
          moe_cluster_parallel_size: int | None = None,
          +
          moe_tensor_parallel_size: int | None = None,
          +
          moe_expert_parallel_size: int | None = None,
          +
          enable_attention_dp: bool = False,
          +
          cp_config: dict | None = <factory>,
          +
          load_format: str | ~tensorrt_llm.llmapi.llm_args.LoadFormat = LoadFormat.AUTO,
          +
          enable_lora: bool = False,
          +
          max_lora_rank: int | None = None,
          +
          max_loras: int = 4,
          +
          max_cpu_loras: int = 4,
          +
          lora_config: ~tensorrt_llm.lora_manager.LoraConfig | None = None,
          +
          enable_prompt_adapter: bool = False,
          +
          max_prompt_adapter_token: int = 0,
          +
          quant_config: ~tensorrt_llm.models.modeling_utils.QuantConfig | None = None,
          +
          kv_cache_config: ~tensorrt_llm.llmapi.llm_args.KvCacheConfig = <factory>,
          +
          enable_chunked_prefill: bool = False,
          +
          guided_decoding_backend: str | None = None,
          +
          batched_logits_processor: object | None = None,
          +
          iter_stats_max_iterations: int | None = None,
          +
          request_stats_max_iterations: int | None = None,
          +
          peft_cache_config: ~tensorrt_llm.llmapi.llm_args.PeftCacheConfig | None = None,
          +
          scheduler_config: ~tensorrt_llm.llmapi.llm_args.SchedulerConfig = <factory>,
          +
          cache_transceiver_config: ~tensorrt_llm.llmapi.llm_args.CacheTransceiverConfig | None = None,
          +
          speculative_config: ~tensorrt_llm.llmapi.llm_args.LookaheadDecodingConfig | ~tensorrt_llm.llmapi.llm_args.MedusaDecodingConfig | ~tensorrt_llm.llmapi.llm_args.EagleDecodingConfig | ~tensorrt_llm.llmapi.llm_args.MTPDecodingConfig | ~tensorrt_llm.llmapi.llm_args.NGramDecodingConfig | None = None,
          +
          batching_type: ~tensorrt_llm.llmapi.llm_args.BatchingType | None = None,
          +
          normalize_log_probs: bool = False,
          +
          max_batch_size: int | None = None,
          +
          max_input_len: int = 1024,
          +
          max_seq_len: int | None = None,
          +
          max_beam_width: int = 1,
          +
          max_num_tokens: int | None = None,
          +
          backend: str | None = None,
          +
          gather_generation_logits: bool = False,
          +
          _num_postprocess_workers: int = 0,
          +
          _postprocess_tokenizer_dir: str | None = None,
          +
          _reasoning_parser: str | None = None,
          +
          decoding_config: object | None = None,
          +
          _mpi_session: object | None = None,
          +
          build_config: object | None = None,
          +
          use_cuda_graph: bool = False,
          +
          cuda_graph_batch_sizes: ~typing.List[int] | None = None,
          +
          cuda_graph_max_batch_size: int = 0,
          +
          cuda_graph_padding_enabled: bool = False,
          +
          disable_overlap_scheduler: bool = False,
          +
          moe_max_num_tokens: int | None = None,
          +
          moe_load_balancer: object | str | None = None,
          +
          attn_backend: str = 'TRTLLM',
          +
          moe_backend: str = 'CUTLASS',
          +
          mixed_sampler: bool = False,
          +
          enable_trtllm_sampler: bool = False,
          +
          kv_cache_dtype: str = 'auto',
          +
          use_kv_cache: bool = True,
          +
          enable_iter_perf_stats: bool = False,
          +
          enable_iter_req_stats: bool = False,
          +
          print_iter_log: bool = False,
          +
          torch_compile_enabled: bool = False,
          +
          torch_compile_fullgraph: bool = True,
          +
          torch_compile_inductor_enabled: bool = False,
          +
          torch_compile_piecewise_cuda_graph: bool = False,
          +
          torch_compile_enable_userbuffers: bool = True,
          +
          autotuner_enabled: bool = True,
          +
          enable_layerwise_nvtx_marker: bool = False,
          +
          auto_deploy_config: object | None = None,
          +
          enable_min_latency: bool = False,
          +
          **extra_data: ~typing.Any,
          +
          + +)[source]#
          +

          Bases: BaseLlmArgs

          +
          +
          +field attn_backend: str = 'TRTLLM'#
          +

          Attention backend to use.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field auto_deploy_config: object | None = None#
          +

          Auto deploy config.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field autotuner_enabled: bool = True#
          +

          Enable autotuner only when torch compile is enabled.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field build_config: object | None = None#
          +

          Build config.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +validator convert_load_format  »  load_format[source]#
          +
          + +
          +
          +field cuda_graph_batch_sizes: List[int] | None = None#
          +

          List of batch sizes to create CUDA graphs for.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field cuda_graph_max_batch_size: int = 0#
          +

          Maximum batch size for CUDA graphs.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +

          • validate_cuda_graph_max_batch_size

            • +
          +
          +
          +
          + +
          +
          +field cuda_graph_padding_enabled: bool = False#
          +

          If true, batches are rounded up to the nearest cuda_graph_batch_size. This is usually a net win for performance.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +decoding_config: object | None#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +field disable_overlap_scheduler: bool = False#
          +

          Disable the overlap scheduler.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field enable_iter_perf_stats: bool = False#
          +

          Enable iteration performance statistics.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field enable_iter_req_stats: bool = False#
          +

          If true, enables per request stats per iteration. Must also set enable_iter_perf_stats to true to get request stats.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field enable_layerwise_nvtx_marker: bool = False#
          +

          If true, enable layerwise nvtx marker.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field enable_min_latency: bool = False#
          +

          If true, enable min-latency mode. Currently only used for Llama4.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field enable_trtllm_sampler: bool = False#
          +

          If true, will use the TRTLLM sampler instead of the PyTorch sampler. The TRTLLM sampler has a wide coverage of sampling strategies.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +property extra_resource_managers: Dict[str, object]#
          +
          + +
          +
          +get_pytorch_backend_config() → PyTorchConfig[source]#
          +
          + +
          +
          +field kv_cache_dtype: str = 'auto'#
          +

          Data type for KV cache.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field load_format: str | LoadFormat = LoadFormat.AUTO#
          +

          How to load the model weights. By default, detect the weight type from the model checkpoint.

          +
          +
          Validated by:
          +
            +

          • convert_load_format

            • +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +max_cpu_loras: int#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +max_lora_rank: int | None#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +max_loras: int#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +field mixed_sampler: bool = False#
          +

          If true, will iterate over sampling_params of each request and use the corresponding sampling strategy, e.g. top-k, top-p, etc.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'extra': 'allow'}#
          +

          Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

          +
          + +
          +
          +model_post_init(_TorchLlmArgs__context)[source]#
          +

          Override this method to perform additional initialization after __init__ and model_construct. +This is useful if you want to do some validation that requires the entire model to be initialized.

          +
          + +
          +
          +field moe_backend: str = 'CUTLASS'#
          +

          MoE backend to use.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field moe_load_balancer: object | str | None = None#
          +

          Configuration for MoE load balancing.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field moe_max_num_tokens: int | None = None#
          +

          If set, at most moe_max_num_tokens tokens will be sent to torch.ops.trtllm.fused_moe at the same time. If the number of tokens exceeds moe_max_num_tokens, the input tensors will be split into chunks and a for loop will be used.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field print_iter_log: bool = False#
          +

          Print iteration logs.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field torch_compile_enable_userbuffers: bool = True#
          +

          When torch compile is enabled, userbuffers is enabled by default.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field torch_compile_enabled: bool = False#
          +

          Enable torch.compile optimization.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field torch_compile_fullgraph: bool = True#
          +

          Enable full graph compilation in torch.compile.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field torch_compile_inductor_enabled: bool = False#
          +

          Enable inductor backend in torch.compile.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field torch_compile_piecewise_cuda_graph: bool = False#
          +

          Enable piecewise CUDA graph in torch.compile.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field use_cuda_graph: bool = False#
          +

          If true, use CUDA graphs for decoding. CUDA graphs are only created for the batch sizes in cuda_graph_batch_sizes, and are enabled for batches that consist of decoding requests only (the reason is that it’s hard to capture a single graph with prefill requests since the input shapes are a function of the sequence lengths). Note that each CUDA graph can use up to 200 MB of extra memory.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +field use_kv_cache: bool = True#
          +

          Whether to use KV cache.

          +
          +
          Validated by:
          +
            +

          • validate_cuda_graph_config

            • +
          +
          +
          +
          + +
          +
          +validator validate_cuda_graph_config  »  all fields[source]#
          +

          Validate CUDA graph configuration.

          +

          Ensures that: +1. If cuda_graph_batch_sizes is provided, cuda_graph_max_batch_size must be 0 +2. If cuda_graph_batch_sizes is not provided, it is generated based on cuda_graph_max_batch_size +3. If both are provided, cuda_graph_batch_sizes must match the generated values

          +
          + +
          +
          +validator validate_cuda_graph_max_batch_size  »  cuda_graph_max_batch_size[source]#
          +

          Validate cuda_graph_max_batch_size is non-negative.

          +
          + +
          + +
          +
          +class tensorrt_llm.llmapi.TrtLlmArgs( + +
          +
          *,
          +
          model: str | ~pathlib.Path,
          +
          tokenizer: str | ~pathlib.Path | ~transformers.tokenization_utils_base.PreTrainedTokenizerBase | ~tensorrt_llm.llmapi.tokenizer.TokenizerBase | None = None,
          +
          tokenizer_mode: ~typing.Literal['auto',
          +
          'slow'] = 'auto',
          +
          skip_tokenizer_init: bool = False,
          +
          trust_remote_code: bool = False,
          +
          tensor_parallel_size: int = 1,
          +
          dtype: str = 'auto',
          +
          revision: str | None = None,
          +
          tokenizer_revision: str | None = None,
          +
          pipeline_parallel_size: int = 1,
          +
          context_parallel_size: int = 1,
          +
          gpus_per_node: int | None = None,
          +
          moe_cluster_parallel_size: int | None = None,
          +
          moe_tensor_parallel_size: int | None = None,
          +
          moe_expert_parallel_size: int | None = None,
          +
          enable_attention_dp: bool = False,
          +
          cp_config: dict | None = <factory>,
          +
          load_format: ~typing.Literal['auto',
          +
          'dummy'] = 'auto',
          +
          enable_lora: bool = False,
          +
          max_lora_rank: int | None = None,
          +
          max_loras: int = 4,
          +
          max_cpu_loras: int = 4,
          +
          lora_config: ~tensorrt_llm.lora_manager.LoraConfig | None = None,
          +
          enable_prompt_adapter: bool = False,
          +
          max_prompt_adapter_token: int = 0,
          +
          quant_config: ~tensorrt_llm.models.modeling_utils.QuantConfig | None = None,
          +
          kv_cache_config: ~tensorrt_llm.llmapi.llm_args.KvCacheConfig = <factory>,
          +
          enable_chunked_prefill: bool = False,
          +
          guided_decoding_backend: str | None = None,
          +
          batched_logits_processor: object | None = None,
          +
          iter_stats_max_iterations: int | None = None,
          +
          request_stats_max_iterations: int | None = None,
          +
          peft_cache_config: ~tensorrt_llm.llmapi.llm_args.PeftCacheConfig | None = None,
          +
          scheduler_config: ~tensorrt_llm.llmapi.llm_args.SchedulerConfig = <factory>,
          +
          cache_transceiver_config: ~tensorrt_llm.llmapi.llm_args.CacheTransceiverConfig | None = None,
          +
          speculative_config: ~tensorrt_llm.llmapi.llm_args.LookaheadDecodingConfig | ~tensorrt_llm.llmapi.llm_args.MedusaDecodingConfig | ~tensorrt_llm.llmapi.llm_args.EagleDecodingConfig | ~tensorrt_llm.llmapi.llm_args.MTPDecodingConfig | ~tensorrt_llm.llmapi.llm_args.NGramDecodingConfig | None = None,
          +
          batching_type: ~tensorrt_llm.llmapi.llm_args.BatchingType | None = None,
          +
          normalize_log_probs: bool = False,
          +
          max_batch_size: int | None = None,
          +
          max_input_len: int = 1024,
          +
          max_seq_len: int | None = None,
          +
          max_beam_width: int = 1,
          +
          max_num_tokens: int | None = None,
          +
          backend: str | None = None,
          +
          gather_generation_logits: bool = False,
          +
          _num_postprocess_workers: int = 0,
          +
          _postprocess_tokenizer_dir: str | None = None,
          +
          _reasoning_parser: str | None = None,
          +
          decoding_config: object | None = None,
          +
          _mpi_session: object | None = None,
          +
          auto_parallel: bool = False,
          +
          auto_parallel_world_size: int | None = None,
          +
          enable_tqdm: bool = False,
          +
          build_config: object | None = None,
          +
          workspace: str | None = None,
          +
          enable_build_cache: object = False,
          +
          extended_runtime_perf_knob_config: ~tensorrt_llm.llmapi.llm_args.ExtendedRuntimePerfKnobConfig | None = None,
          +
          calib_config: ~tensorrt_llm.llmapi.llm_args.CalibConfig | None = None,
          +
          embedding_parallel_mode: str = 'SHARDING_ALONG_VOCAB',
          +
          fast_build: bool = False,
          +
          **extra_data: ~typing.Any,
          +
          + +)[source]#
          +

          Bases: BaseLlmArgs

          +
          +
          +auto_parallel: bool#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +property auto_parallel_config: AutoParallelConfig#
          +
          + +
          +
          +auto_parallel_world_size: int | None#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +field build_config: object | None = None#
          +

          Build config.

          +
          + +
          +
          +field calib_config: CalibConfig | None = None#
          +

          Calibration config.

          +
          + +
          +
          +decoding_config: object | None#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +field embedding_parallel_mode: str = 'SHARDING_ALONG_VOCAB'#
          +

          The embedding parallel mode.

          +
          + +
          +
          +field enable_build_cache: object = False#
          +

          Enable build cache.

          +
          + +
          +
          +field enable_tqdm: bool = False#
          +

          Enable tqdm for progress bar.

          +
          + +
          +
          +field extended_runtime_perf_knob_config: ExtendedRuntimePerfKnobConfig | None = None#
          +

          Extended runtime perf knob config.

          +
          + +
          +
          +field fast_build: bool = False#
          +

          Enable fast build.

          +
          + +
          +
          +max_cpu_loras: int#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +max_lora_rank: int | None#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +max_loras: int#
          +

          Read-only data descriptor used to emit a runtime deprecation warning before accessing a deprecated field.

          +
          +
          +msg#
          +

          The deprecation message to be emitted.

          +
          + +
          +
          +wrapped_property#
          +

          The property instance if the deprecated field is a computed field, or None.

          +
          + +
          +
          +field_name#
          +

          The name of the field being deprecated.

          +
          + +
          + +
          +
          +model_config: ClassVar[ConfigDict] = {'arbitrary_types_allowed': True, 'extra': 'allow'}#
          +

          Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

          +
          + +
          +
          +model_post_init(_TrtLlmArgs__context)[source]#
          +

          Override this method to perform additional initialization after __init__ and model_construct. +This is useful if you want to do some validation that requires the entire model to be initialized.

          +
          + +
          +
          +field workspace: str | None = None#
          +

          The workspace for the model.

          +
          + +
          + @@ -3371,7 +4363,9 @@ changed, you should remove the caches manually.

        • __init__()
        • decode_duration_ms
        • decode_retention_priority
          • +
        • directory
        • token_range_retention_configs
          • +
        • transfer_mode
        • LookaheadDecodingConfig
        • MedusaDecodingConfig @@ -3667,6 +4785,15 @@ changed, you should remove the caches manually.

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/objects.inv b/latest/objects.inv index 49f74f2b4e..8a655c18d0 100644 Binary files a/latest/objects.inv and b/latest/objects.inv differ diff --git a/latest/overview.html b/latest/overview.html index 2e9115ebaa..fbb0cb079b 100644 --- a/latest/overview.html +++ b/latest/overview.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -698,6 +702,15 @@ Certain limitations might apply. Refer to the +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/perf-analysis.html b/latest/performance/perf-analysis.html index 679ec35c22..527a17526d 100644 --- a/latest/performance/perf-analysis.html +++ b/latest/performance/perf-analysis.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -757,6 +761,15 @@ python3 benchmarks/cpp/prepare_dataset.py

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/perf-benchmarking.html b/latest/performance/perf-benchmarking.html index 5c31b3a82f..1c7fccb60e 100644 --- a/latest/performance/perf-benchmarking.html +++ b/latest/performance/perf-benchmarking.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -1166,8 +1170,7 @@ follow when a checkpoint does not specify a KV cache quantization algorithm:

          If you would like to force the KV cache quantizaton, you can specify the following in the YAML file to force the precision when the checkpoint precision is null:

          -
          pytorch_backend_config:
-  kv_cache_dtype: "fp8"
+
          kv_cache_dtype: "fp8"
 
          
 
          
 
          
@@ -1537,6 +1540,15 @@ The choices are specified with a YAML file like the following example (
 
          
       
+        
          
+
          
+  
+  
          Last updated on June 03, 2025.
          
+  
+  
          This page is generated by TensorRT-LLM commit 9ae2ce6.
          
+  
+
          
+
          diff --git a/latest/performance/perf-overview.html b/latest/performance/perf-overview.html index efeb575c54..6b5d83fde2 100644 --- a/latest/performance/perf-overview.html +++ b/latest/performance/perf-overview.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -1223,11 +1227,9 @@ a model name (HuggingFace reference or path to a local model), a llm_options.yml

          -
           pytorch_backend_config:
-  enable_overlap_scheduler: true
-  use_cuda_graph: true
-  cuda_graph_padding_enabled: true
-  cuda_graph_batch_sizes:
+
          use_cuda_graph: true
+cuda_graph_padding_enabled: true
+cuda_graph_batch_sizes:
   - 1
   - 2
   - 4
@@ -1462,6 +1464,15 @@ using the --kv_cach
   
          
 
          
       
+        
          
+
          
+  
+  
          Last updated on June 03, 2025.
          
+  
+  
          This page is generated by TensorRT-LLM commit 9ae2ce6.
          
+  
+
          
+      
     
          
   
   
diff --git a/latest/performance/performance-tuning-guide/benchmarking-default-performance.html b/latest/performance/performance-tuning-guide/benchmarking-default-performance.html
index 23b85795d8..59ab78800c 100644
--- a/latest/performance/performance-tuning-guide/benchmarking-default-performance.html
+++ b/latest/performance/performance-tuning-guide/benchmarking-default-performance.html
@@ -51,7 +51,7 @@
     
@@ -63,7 +63,7 @@
 
   
   
-  
+  
 
 
   
@@ -336,6 +336,7 @@
 
        • Generate Text Asynchronously
          • 
 
        • Distributed LLM Generation
          • 
 
        • Control generated text using logits processor
          • 
+
        • Generate Text Using Eagle2 Decoding
          • 
 
        • Get KV Cache Events
          • 
 
        • Generate Text Using Lookahead Decoding
          • 
 
        • Generation with Quantization
          • 
@@ -357,6 +358,7 @@
 
        • Generate Text Asynchronously
          • 
 
        • Distributed LLM Generation
          • 
 
        • Control generated text using logits processor
          • 
+
        • Generate Text Using Eagle2 Decoding
          • 
 
        • Get KV Cache Events
          • 
 
        • Generate Text Using Lookahead Decoding
          • 
 
        • Generation with Quantization
          • 
@@ -421,6 +423,7 @@
 
        • Graph Rewriting Module
          • 
 
        • Run gpt-2b + LoRA using Executor / cpp runtime
          • 
 
        • Expert Parallelism in TensorRT-LLM
          • 
+
        • KV Cache Management: Pools, Blocks, and Events
          • 
 
        • KV cache reuse
          • 
 
        • Speculative Sampling
          • 
 
        • Disaggregated-Service (experimental)
          • 
@@ -455,6 +458,7 @@
 
        • Speed up inference with SOTA quantization techniques in TRT-LLM
          • 
 
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
          • 
 
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • 
+
        • DeepSeek R1 MTP Implementation and Optimization
          •
          @@ -908,6 +912,15 @@ P99: 1.00

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/deciding-model-sharding-strategy.html b/latest/performance/performance-tuning-guide/deciding-model-sharding-strategy.html index 093ab20674..f93cbdd593 100644 --- a/latest/performance/performance-tuning-guide/deciding-model-sharding-strategy.html +++ b/latest/performance/performance-tuning-guide/deciding-model-sharding-strategy.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -687,6 +691,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/fp8-quantization.html b/latest/performance/performance-tuning-guide/fp8-quantization.html index 8258b55aa9..990e8fb4bb 100644 --- a/latest/performance/performance-tuning-guide/fp8-quantization.html +++ b/latest/performance/performance-tuning-guide/fp8-quantization.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -1019,6 +1023,15 @@ accuracy loss is unacceptable.

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/index.html b/latest/performance/performance-tuning-guide/index.html index 2f6efd3851..6da81fa5ef 100644 --- a/latest/performance/performance-tuning-guide/index.html +++ b/latest/performance/performance-tuning-guide/index.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -678,6 +682,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.html b/latest/performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.html index 30d4b2e723..84b30626d9 100644 --- a/latest/performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.html +++ b/latest/performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -869,6 +873,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/useful-build-time-flags.html b/latest/performance/performance-tuning-guide/useful-build-time-flags.html index 129ae2cdf8..d4bd801cd6 100644 --- a/latest/performance/performance-tuning-guide/useful-build-time-flags.html +++ b/latest/performance/performance-tuning-guide/useful-build-time-flags.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -932,6 +936,15 @@ This can be enabled via the LLM-API as such

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/performance/performance-tuning-guide/useful-runtime-flags.html b/latest/performance/performance-tuning-guide/useful-runtime-flags.html index ac6bcf3b92..d4d2715d26 100644 --- a/latest/performance/performance-tuning-guide/useful-runtime-flags.html +++ b/latest/performance/performance-tuning-guide/useful-runtime-flags.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -855,6 +859,15 @@ via KVCacheConfig +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/py-modindex.html b/latest/py-modindex.html index 5491802114..a871415b16 100644 --- a/latest/py-modindex.html +++ b/latest/py-modindex.html @@ -50,7 +50,7 @@ @@ -61,7 +61,7 @@ - + @@ -332,6 +332,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -353,6 +354,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -417,6 +419,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -451,6 +454,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -674,6 +678,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/python-api/tensorrt_llm.functional.html b/latest/python-api/tensorrt_llm.functional.html index d91f98cab4..65f4ac9f8e 100644 --- a/latest/python-api/tensorrt_llm.functional.html +++ b/latest/python-api/tensorrt_llm.functional.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -634,6 +638,11 @@ AUTO = 3#
          +
          +
          +LOWPRECISION = 6#
          +
          +
          MIN_LATENCY = 1#
          @@ -6638,6 +6647,7 @@ function creates a constant tensor.

        • AllReduceStrategy @@ -2601,6 +2605,15 @@ the number of tokens used for each task, should be equal to prompt_embedding_tab

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/python-api/tensorrt_llm.models.html b/latest/python-api/tensorrt_llm.models.html index 4603150395..994310abbd 100644 --- a/latest/python-api/tensorrt_llm.models.html +++ b/latest/python-api/tensorrt_llm.models.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -3031,6 +3035,11 @@ ranges of the dimensions of when using TRT dynamic shapes.

          MEDUSA = 4#
          +
          +
          +NGRAM = 64#
          +
          +
          NONE = 1#
          @@ -3466,6 +3475,7 @@ ranges of the dimensions of when using TRT dynamic shapes.

        • EXPLICIT_DRAFT_TOKENS
        • LOOKAHEAD_DECODING
        • MEDUSA
          • +
        • NGRAM
        • NONE
        • from_arguments()
          • @@ -3567,6 +3577,15 @@ ranges of the dimensions of when using TRT dynamic shapes.

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/python-api/tensorrt_llm.plugin.html b/latest/python-api/tensorrt_llm.plugin.html index fc3f02c8d1..3ed44e3115 100644 --- a/latest/python-api/tensorrt_llm.plugin.html +++ b/latest/python-api/tensorrt_llm.plugin.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -691,6 +695,15 @@ migrated to the centralized building script tensorrt_llm/commands/build.py

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/python-api/tensorrt_llm.quantization.html b/latest/python-api/tensorrt_llm.quantization.html index c8639836ca..eb023c5228 100644 --- a/latest/python-api/tensorrt_llm.quantization.html +++ b/latest/python-api/tensorrt_llm.quantization.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -729,6 +733,15 @@ the quantized model as TRT-LLM checkpoint

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/python-api/tensorrt_llm.runtime.html b/latest/python-api/tensorrt_llm.runtime.html index a784554226..c4f6e49965 100644 --- a/latest/python-api/tensorrt_llm.runtime.html +++ b/latest/python-api/tensorrt_llm.runtime.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -3271,6 +3275,15 @@ For example, word_dict[2] = [” I am happy”, “ I am sad”].

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/quick-start-guide.html b/latest/quick-start-guide.html index c7fa260d17..597747797c 100644 --- a/latest/quick-start-guide.html +++ b/latest/quick-start-guide.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -831,6 +835,15 @@ The model definition is a minimal example that shows some of the optimizations a

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/reference/memory.html b/latest/reference/memory.html index 18d5ce4a34..299aadd787 100644 --- a/latest/reference/memory.html +++ b/latest/reference/memory.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -778,6 +782,15 @@ Here some explanations on how these values affect the memory:

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/reference/precision.html b/latest/reference/precision.html index c79d59135e..9952239e75 100644 --- a/latest/reference/precision.html +++ b/latest/reference/precision.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -1274,6 +1278,15 @@ are:

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/reference/support-matrix.html b/latest/reference/support-matrix.html index 2c1595d05c..667a4a5efa 100644 --- a/latest/reference/support-matrix.html +++ b/latest/reference/support-matrix.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -934,6 +938,15 @@ In addition, older architectures can have limitations for newer software release

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/reference/troubleshooting.html b/latest/reference/troubleshooting.html index 5356d7288f..3137b6b16d 100644 --- a/latest/reference/troubleshooting.html +++ b/latest/reference/troubleshooting.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -964,6 +968,15 @@ dedicated MPI environment, not the one provided by your Slurm allocation.

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/release-notes.html b/latest/release-notes.html index 515c325086..6dc54b2080 100644 --- a/latest/release-notes.html +++ b/latest/release-notes.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -2041,6 +2045,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/search.html b/latest/search.html index 2537c79245..44d45c0d6e 100644 --- a/latest/search.html +++ b/latest/search.html @@ -51,7 +51,7 @@ @@ -69,7 +69,7 @@ - + @@ -338,6 +338,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -359,6 +360,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -423,6 +425,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -457,6 +460,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -619,6 +623,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/searchindex.js b/latest/searchindex.js index e8012775e0..16ebaf26ff 100644 --- a/latest/searchindex.js +++ b/latest/searchindex.js @@ -1 +1 @@ -Search.setIndex({"alltitles": {"1. Download TensorRT-LLM": [[18, "download-tensorrt-llm"]], "1. Weights size": [[84, "weights-size"]], "2. Activation size": [[84, "activation-size"]], "2. Download the DeepSeek R1 models": [[18, "download-the-deepseek-r1-models"]], "3. Build and run TensorRT-LLM container": [[18, "build-and-run-tensorrt-llm-container"]], "3. I/O tensors": [[84, "i-o-tensors"]], "3.1 Runtime and decoder buffers except KV cache tensor": [[84, "runtime-and-decoder-buffers-except-kv-cache-tensor"]], "3.2 KV cache tensor": [[84, "kv-cache-tensor"]], "4. Compile and Install TensorRT-LLM": [[18, "compile-and-install-tensorrt-llm"]], "5. Optional: Tune GPU clocks": [[18, "optional-tune-gpu-clocks"]], "6. Dataset preparation": [[18, "dataset-preparation"]], "@record_signature to Decorate Functionals Requiring FLayerInfo": [[7, "record-signature-to-decorate-functionals-requiring-flayerinfo"]], "ALiBi": [[5, "alibi"]], "API": [[3, "api"]], "API Changes": [[11, "api-changes"], [88, "api-changes"], [88, "id9"], [88, "id14"], [88, "id19"], [88, "id24"], [88, "id31"], [88, "id36"], [88, "id42"], [88, "id48"], [88, "id54"]], "API Introduction": [[64, null]], "API Reference": [[65, null]], "AWQ Quantization Scaling Factors": [[13, "awq-quantization-scaling-factors"]], "About": [[26, "about"]], "About Speculative Sampling": [[10, "about-speculative-sampling"]], "About TensorRT-LLM": [[66, "about-tensorrt-llm"]], "Accuracy": [[23, "accuracy"]], "Acknowledgment": [[24, "acknowledgment"]], "Activation": [[78, "module-tensorrt_llm.layers.activation"]], "Adding a Model": [[12, null]], "Adding a New Model in PyTorch Backend": [[90, null]], "Advanced": [[59, null]], "Announcements": [[88, "announcements"], [88, "id52"]], "Architecture": [[59, null]], "Architecture Ovewiew": [[91, null]], "Asyncio-Based Generation": [[32, "asyncio-based-generation"]], "Attention": [[78, "module-tensorrt_llm.layers.attention"], [92, null]], "Attention Backends": [[92, "attention-backends"]], "Attention Kernel": [[24, "attention-kernel"]], "Attention Weights": [[13, "attention-weights"]], "Auto parallel arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-auto-parallel-arguments"]], "Automatic Parallelism with LLM": [[38, null]], "Autoregressive MTP Layers": [[24, "autoregressive-mtp-layers"]], "B200 max-throughput": [[18, "b200-max-throughput"]], "B200 min-latency": [[18, "b200-min-latency"]], "Background": [[24, "background"]], "Beam-Search": [[5, "beam-search"]], "Before Benchmarking": [[68, "before-benchmarking"]], "Before You Begin: TensorRT-LLM LLM-API": [[70, "before-you-begin-tensorrt-llm-llm-api"]], "Benchmark": [[18, "benchmark"], [23, "benchmark"], [26, "benchmark"]], "Benchmarking Default Performance": [[70, null]], "Benchmarking a non-Medusa Low Latency Engine": [[68, "benchmarking-a-non-medusa-low-latency-engine"]], "Benchmarking with trtllm-bench": [[70, "benchmarking-with-trtllm-bench"]], "Benchmarks": [[2, "benchmarks"]], "Best practices to choose the right quantization methods": [[23, "best-practices-to-choose-the-right-quantization-methods"]], "Boost settings": [[68, "boost-settings"]], "Build APIs": [[17, "build-apis"]], "Build Checkpoint into TensorRT Engine": [[13, "build-checkpoint-into-tensorrt-engine"]], "Build Configuration": [[32, "build-configuration"]], "Build TensorRT-LLM": [[60, "build-tensorrt-llm"]], "Build the TensorRT-LLM Docker Image": [[27, null]], "Build the TensorRT-LLM Docker Image and Upload to DockerHub": [[27, "build-the-tensorrt-llm-docker-image-and-upload-to-dockerhub"], [28, "build-the-tensorrt-llm-docker-image-and-upload-to-dockerhub"]], "Building a Benchmark Engine": [[68, "building-a-benchmark-engine"]], "Building a Medusa Low-Latency Engine": [[68, "building-a-medusa-low-latency-engine"]], "Building a TensorRT-LLM Docker Image": [[60, "building-a-tensorrt-llm-docker-image"]], "Building and Saving Engines via CLI": [[70, "building-and-saving-engines-via-cli"]], "Building and Saving the Engine": [[70, "building-and-saving-the-engine"]], "Building from Source Code on Linux": [[60, null]], "Building the Python Bindings for the C++ Runtime": [[60, "building-the-python-bindings-for-the-c-runtime"]], "C++ Executor API Example": [[3, "c-executor-api-example"]], "C++ GPT Runtime": [[6, null]], "C++ runtime": [[84, "c-runtime"], [84, "id1"]], "CLI Tools": [[17, "cli-tools"]], "CUDA Graph & Programmatic Dependent Launch": [[24, "cuda-graph-programmatic-dependent-launch"]], "CUTLASS Backend (default backend)": [[24, "cutlass-backend-default-backend"]], "Capacity Scheduler Policy": [[76, "capacity-scheduler-policy"]], "Cast": [[78, "module-tensorrt_llm.layers.cast"]], "Chat API": [[26, "chat-api"]], "Chunked Context": [[5, "chunked-context"]], "Classical Workflow": [[7, "classical-workflow"]], "Closing": [[19, "closing"], [22, "closing"]], "Collect PyTorch profiler results": [[67, "collect-pytorch-profiler-results"]], "Command Overview": [[69, "command-overview"]], "Common LLM Support": [[66, "common-llm-support"]], "Communication Kernel": [[24, "communication-kernel"]], "Compilation": [[14, "compilation"]], "Compile the Model into a TensorRT Engine": [[83, "compile-the-model-into-a-tensorrt-engine"]], "Completions API": [[26, "completions-api"], [26, "id1"]], "Conclusion": [[72, "conclusion"], [74, "conclusion"], [75, "conclusion"]], "Config": [[13, "config"]], "Configure SSH Key": [[28, "configure-ssh-key"]], "Configure The Executor": [[3, "configure-the-executor"]], "Connect to the Pod": [[28, "connect-to-the-pod"]], "Context Chunking Policy": [[76, "context-chunking-policy"]], "Context Phase": [[5, "context-phase"]], "Context and Generation Phases": [[5, "context-and-generation-phases"]], "Contiguous KV Cache": [[5, "contiguous-kv-cache"]], "Control generated text using logits processor": [[47, null]], "Controlling output with Logits Post-Processor": [[3, "controlling-output-with-logits-post-processor"]], "Conv": [[78, "module-tensorrt_llm.layers.conv"]], "Conversion APIs": [[17, "conversion-apis"]], "Coordinating with NVIDIA Nsight Systems Launch": [[67, "coordinating-with-nvidia-nsight-systems-launch"]], "Coordinating with PyTorch profiler (PyTorch workflow only)": [[67, "coordinating-with-pytorch-profiler-pytorch-workflow-only"]], "Core Models": [[90, "core-models"]], "Create a Pod Template": [[28, "create-a-pod-template"]], "Create a Runpod account": [[28, "create-a-runpod-account"]], "Create the Container": [[60, "create-the-container"]], "Cross Attention": [[5, "cross-attention"]], "Curl Chat Client": [[29, null]], "Curl Chat Client For Multimodal": [[30, null]], "Curl Completion Client": [[31, null]], "Customize KV Cache Manager": [[93, "customize-kv-cache-manager"]], "Customize Your Own Scheduler": [[94, "customize-your-own-scheduler"]], "Debug Execution Errors": [[87, "debug-execution-errors"]], "Debug on E2E Models": [[87, "debug-on-e2e-models"]], "Debug on Unit Tests": [[87, "debug-on-unit-tests"]], "Debugging FAQs": [[2, "debugging-faqs"]], "Deciding Model Sharding Strategy": [[71, null]], "Decoder": [[91, "decoder"]], "Deepseek R1 Reasoning Parser": [[33, null]], "Default Build Behavior": [[68, "default-build-behavior"]], "Dense GEMM optimization": [[24, "dense-gemm-optimization"]], "Deploy with Triton Inference Server": [[83, "deploy-with-triton-inference-server"]], "Deploy with trtllm-serve": [[83, "deploy-with-trtllm-serve"]], "Develop TensorRT-LLM on Runpod": [[28, null]], "Developer Guide": [[89, "developer-guide"]], "Disable Tokenizer": [[32, "disable-tokenizer"]], "Disaggregated-Service (experimental)": [[2, null]], "Distributed LLM Generation": [[45, null]], "DoRA": [[9, "dora"]], "Documentation": [[88, "documentation"], [88, "id28"]], "Draft-Target-Model": [[10, "draft-target-model"]], "EAGLE": [[10, "eagle"]], "Embedding": [[78, "module-tensorrt_llm.layers.embedding"]], "Enable GIL information in NVTX markers": [[67, "enable-gil-information-in-nvtx-markers"]], "Enable garbage collection (GC) NVTX markers": [[67, "enable-garbage-collection-gc-nvtx-markers"]], "Enable kv cache reuse for p-tuning": [[8, "enable-kv-cache-reuse-for-p-tuning"]], "Enable more NVTX markers for debugging": [[67, "enable-more-nvtx-markers-for-debugging"]], "Enable ssh access to the container": [[27, "enable-ssh-access-to-the-container"]], "Enabling GEMM + SwiGLU Fusion": [[72, "enabling-gemm-swiglu-fusion"]], "Enabling GEMM Plugin": [[75, "enabling-gemm-plugin"]], "Enabling Low Latency GEMM plugin": [[72, "enabling-low-latency-gemm-plugin"]], "Enabling Paged Context Attention": [[75, "enabling-paged-context-attention"]], "Enabling Quantization": [[72, "enabling-quantization"]], "Enabling Quantized KV Cache": [[72, "enabling-quantized-kv-cache"]], "Enabling Reduce Norm Fusion Plugin": [[75, "enabling-reduce-norm-fusion-plugin"]], "Enabling Reduce Norm Fusion with User Buffers": [[72, "enabling-reduce-norm-fusion-with-user-buffers"]], "Enabling building with multiple profiles": [[75, "enabling-building-with-multiple-profiles"]], "Environment Variables": [[2, "environment-variables"]], "Everything in One Diagram": [[24, "everything-in-one-diagram"]], "Example": [[2, "example"], [13, "example"]], "Example LoRA tensors": [[9, "example-lora-tensors"]], "Example of Build Subcommand Output:": [[68, "example-of-build-subcommand-output"]], "Examples": [[14, "examples"], [15, "examples"], [67, "examples"]], "Executor": [[0, null]], "Executor API": [[3, null]], "Expected Result Format": [[18, "expected-result-format"], [18, "id1"], [18, "id2"]], "Expected Results": [[18, "expected-results"]], "Expert Parallelism in TensorRT-LLM": [[4, null]], "Exploring more ISL/OSL combinations": [[18, "exploring-more-isl-osl-combinations"]], "FAQ": [[84, "faq"]], "FLayerInfo for Retrieving High-Level Information for a Functional": [[7, "flayerinfo-for-retrieving-high-level-information-for-a-functional"]], "FP32, FP16 and BF16": [[85, "fp32-fp16-and-bf16"]], "FP4 Models:": [[69, "fp4-models"]], "FP8 (Hopper)": [[85, "fp8-hopper"]], "FP8 Context FMHA": [[5, "fp8-context-fmha"]], "FP8 Models:": [[69, "fp8-models"]], "FP8 Quantization": [[72, null]], "FP8 Quantization Scaling Factors": [[13, "fp8-quantization-scaling-factors"]], "FP8 Support": [[66, "fp8-support"]], "FP8 \u201cBaseline\u201d Performance": [[72, "fp8-baseline-performance"]], "Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100": [[19, null]], "Falcon-180B on a single H200 with INT4 AWQ": [[19, "falcon-180b-on-a-single-h200-with-int4-awq"]], "Feature Descriptions": [[67, "feature-descriptions"]], "Fixed Issues": [[88, "fixed-issues"], [88, "id11"], [88, "id15"], [88, "id21"], [88, "id26"], [88, "id33"], [88, "id38"], [88, "id44"], [88, "id50"], [88, "id56"], [88, "id61"]], "Fully customized": [[15, "fully-customized"]], "Functionals": [[77, null]], "Fuse_A_GEMM": [[24, "fuse-a-gemm"]], "Future Works": [[24, "future-works"]], "Future-Style Generation": [[32, "future-style-generation"]], "GEMM + SwiGLU Fusion in Gated-MLP": [[72, "gemm-swiglu-fusion-in-gated-mlp"]], "GEMM Plugin": [[75, "gemm-plugin"]], "GPTQ and AWQ (W4A16)": [[85, "gptq-and-awq-w4a16"]], "GPU Clock Management": [[68, "gpu-clock-management"]], "Genai Perf Client": [[34, null]], "Genai Perf Client For Multimodal": [[35, null]], "General FAQs": [[2, "general-faqs"]], "Generate Text Asynchronously": [[42, null]], "Generate Text Using Eagle Decoding": [[39, null]], "Generate Text Using Lookahead Decoding": [[48, null]], "Generate Text Using Medusa Decoding": [[49, null]], "Generate Text in Streaming": [[43, null]], "Generate text": [[41, null]], "Generate text with customization": [[44, null]], "Generate text with guided decoding": [[40, null]], "Generate text with multiple LoRA adapters": [[53, null]], "Generation": [[32, "generation"]], "Generation Phase": [[5, "generation-phase"]], "Generation with Quantization": [[54, null]], "Get KV Cache Events": [[46, null]], "Getting Started": [[59, null]], "Graph Rewriting APIs": [[7, "graph-rewriting-apis"]], "Graph Rewriting Module": [[7, null]], "Grouped GEMM": [[24, "grouped-gemm"]], "H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token": [[20, null]], "H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLM": [[21, null]], "H200 max-throughput": [[18, "h200-max-throughput"]], "H200 min-latency": [[18, "h200-min-latency"]], "H200 vs H100": [[21, "h200-vs-h100"]], "Hardware": [[86, "hardware"]], "How the Benchmarker Works": [[68, "how-the-benchmarker-works"]], "How to Enable": [[4, "how-to-enable"]], "How to Think about Model Sharding: Communication is Key": [[71, "how-to-think-about-model-sharding-communication-is-key"]], "How to change Max Batch Size": [[74, "how-to-change-max-batch-size"]], "How to change Max Num Tokens": [[74, "how-to-change-max-num-tokens"]], "How to enable kv cache reuse": [[8, "how-to-enable-kv-cache-reuse"]], "How to get best performance on DeepSeek-R1 in TensorRT-LLM": [[18, null]], "How to reproduce": [[24, "how-to-reproduce"]], "How to set Tensor Parallelism and Pipeline Parallelism": [[71, "how-to-set-tensor-parallelism-and-pipeline-parallelism"]], "Hugging Face Hub": [[64, "hugging-face-hub"]], "INT4 and INT8 Weight-Only (W4A16 and W8A16)": [[85, "int4-and-int8-weight-only-w4a16-and-w8a16"]], "INT8 SmoothQuant (W8A8)": [[85, "int8-smoothquant-w8a8"]], "INT8/FP8 KV Caches": [[5, "int8-fp8-kv-caches"]], "Implement AttentionBackend": [[92, "implement-attentionbackend"]], "Implement AttentionMetadata": [[92, "implement-attentionmetadata"]], "Implement a New Attention Backend": [[92, "implement-a-new-attention-backend"]], "Implementation Configuration": [[24, "implementation-configuration"]], "Important Note": [[5, "important-note"]], "In-Flight Batching and Paged Attention": [[66, "in-flight-batching-and-paged-attention"]], "In-flight Batching": [[5, "in-flight-batching"]], "In-flight Batching with the Triton Inference Server": [[3, "in-flight-batching-with-the-triton-inference-server"]], "Indices and tables": [[59, "indices-and-tables"]], "Inference Endpoints": [[26, "inference-endpoints"]], "Infrastructure Changes": [[88, "infrastructure-changes"], [88, "id4"], [88, "id7"], [88, "id12"], [88, "id16"], [88, "id22"], [88, "id27"], [88, "id34"], [88, "id39"], [88, "id45"]], "Infrastructure changes": [[88, "id51"]], "Input QKV tensor": [[5, "input-qkv-tensor"]], "Installation": [[59, null]], "Installation Errors": [[87, "installation-errors"]], "Installing on Grace Hopper": [[61, null]], "Installing on Linux": [[62, null]], "Interfaces": [[93, "interfaces"]], "Internal Components": [[6, "internal-components"]], "Introduction": [[90, "introduction"]], "KV Cache": [[5, "kv-cache"]], "KV Cache Manager": [[93, null]], "KV Cache Manager Introduction": [[93, "kv-cache-manager-introduction"]], "KV Cache Quantization Scaling Factors": [[13, "kv-cache-quantization-scaling-factors"]], "KV cache reuse": [[8, null]], "KVCacheManager": [[91, "kvcachemanager"]], "Kernel Level optimizations": [[24, "kernel-level-optimizations"]], "Kernel fusion": [[24, "kernel-fusion"]], "Key Components": [[89, "key-components"]], "Key Features": [[63, null]], "Key Features and Enhancements": [[88, "key-features-and-enhancements"], [88, "id2"], [88, "id3"], [88, "id5"], [88, "id8"], [88, "id13"], [88, "id18"], [88, "id23"], [88, "id30"], [88, "id35"], [88, "id41"], [88, "id47"], [88, "id53"], [88, "id57"], [88, "id59"]], "Key Optimizations": [[24, "key-optimizations"]], "Known Issues": [[84, "known-issues"], [88, "known-issues"], [88, "id6"], [88, "id10"], [88, "id17"], [88, "id29"], [88, "id40"], [88, "id46"], [88, "id62"], [89, "known-issues"]], "Known Limitations": [[60, "known-limitations"]], "LLM API": [[83, "llm-api"]], "LLM API Examples": [[36, null]], "LLM Common Customizations": [[32, null]], "LLM Examples": [[37, null]], "LLM Examples Introduction": [[36, null]], "LLM Models": [[86, "llm-models"]], "Latest GPU Support": [[66, "latest-gpu-support"]], "Latest HBM Memory": [[21, "latest-hbm-memory"]], "LayerNorm Weights": [[13, "layernorm-weights"]], "Layers": [[78, null]], "Limitations": [[10, "limitations"], [88, "limitations"]], "Limitations and Caveats": [[68, "limitations-and-caveats"]], "Linear": [[78, "module-tensorrt_llm.layers.linear"]], "Linking with the TensorRT-LLM C++ Runtime": [[60, "linking-with-the-tensorrt-llm-c-runtime"]], "Llama 3.1 405B": [[14, "llama-3-1-405b"]], "Llama 3.1 405B FP4": [[69, "llama-3-1-405b-fp4"]], "Llama 3.1 405B FP8": [[69, "llama-3-1-405b-fp8"]], "Llama 3.1 70B": [[14, "llama-3-1-70b"]], "Llama 3.1 70B FP8": [[69, "llama-3-1-70b-fp8"]], "Llama 3.1 8B FP8": [[69, "llama-3-1-8b-fp8"]], "Llama 3.3 70B FP4": [[69, "llama-3-3-70b-fp4"]], "Llama-70B on H200 up to 2.4x increased throughput with XQA within same latency budget": [[22, "llama-70b-on-h200-up-to-2-4x-increased-throughput-with-xqa-within-same-latency-budget"]], "Llama-70B on H200 up to 6.7x A100": [[19, "llama-70b-on-h200-up-to-6-7x-a100"]], "Llm Mgmn Llm Distributed": [[50, null]], "Llm Mgmn Trtllm Bench": [[51, null]], "Llm Mgmn Trtllm Serve": [[52, null]], "LoRA Module id mapping": [[9, "lora-module-id-mapping"]], "LoRA arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-lora-arguments"]], "LoRA tensor format details": [[9, "lora-tensor-format-details"]], "LoRA with tensor parallel": [[9, "lora-with-tensor-parallel"]], "Loading function": [[15, "loading-function"]], "Local Hugging Face Models": [[64, "local-hugging-face-models"]], "Local TensorRT-LLM Engine": [[64, "local-tensorrt-llm-engine"]], "Logits arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-logits-arguments"]], "Lookahead Decoding": [[10, "lookahead-decoding"]], "LoraCache configuration": [[9, "loracache-configuration"]], "Low Latency Benchmark": [[68, "low-latency-benchmark"]], "Low Latency GEMM Plugin": [[72, "low-latency-gemm-plugin"]], "Low Latency TensorRT-LLM Engine for Llama-3 70B": [[68, "low-latency-tensorrt-llm-engine-for-llama-3-70b"]], "MLP": [[78, "module-tensorrt_llm.layers.mlp"]], "MLP Weights": [[13, "mlp-weights"]], "MLPerf on H100 with FP8": [[20, "mlperf-on-h100-with-fp8"]], "MTP": [[24, "mtp"]], "Make Evaluation": [[13, "make-evaluation"]], "Mark Tensors As Output": [[3, "mark-tensors-as-output"]], "Max Throughput Benchmark": [[68, "max-throughput-benchmark"]], "Max Tokens in Paged KV Cache and KV Cache Free GPU Memory Fraction": [[76, "max-tokens-in-paged-kv-cache-and-kv-cache-free-gpu-memory-fraction"]], "Maximum Attention Window Size": [[76, "maximum-attention-window-size"]], "Medusa": [[10, "medusa"]], "Medusa Tree": [[10, "medusa-tree"]], "Memory Usage of TensorRT-LLM": [[84, null]], "Memory pool": [[84, "memory-pool"]], "Metrics Endpoint": [[26, "metrics-endpoint"]], "Mixed ETP": [[24, "mixed-etp"]], "Mixture of Experts (MoE)": [[4, "mixture-of-experts-moe"]], "Model Architecture": [[24, "model-architecture"]], "Model Configuration": [[6, "model-configuration"], [90, "model-configuration"]], "Model Definition": [[14, null], [90, "model-definition"]], "Model Definition API": [[83, "model-definition-api"]], "Model Engine": [[14, "model-engine"], [91, "model-engine"]], "Model Preparation": [[64, "model-preparation"]], "Model Registration": [[90, "model-registration"]], "Model Updates": [[88, "model-updates"], [88, "id20"], [88, "id25"], [88, "id32"], [88, "id37"], [88, "id43"], [88, "id49"], [88, "id55"], [88, "id58"], [88, "id60"]], "Model Weights": [[16, "model-weights"]], "Models": [[79, null]], "Models (PyTorch Backend)": [[86, "models-pytorch-backend"]], "Models (TensorRT Backend)": [[86, "models-tensorrt-backend"]], "Models with customized key names": [[15, "models-with-customized-key-names"]], "Models with customized weight layout": [[15, "models-with-customized-weight-layout"]], "Multi-GPU Multi-Node Inference": [[66, "multi-gpu-multi-node-inference"]], "Multi-GPU and Multi-Node Support": [[14, "multi-gpu-and-multi-node-support"]], "Multi-Head, Multi-Query, and Group-Query Attention": [[5, null]], "Multi-Modal Models 3": [[86, "multi-modal-models"]], "Multi-node Serving with Slurm": [[26, "multi-node-serving-with-slurm"]], "Multi-streams": [[24, "multi-streams"]], "Multimodal Serving": [[26, "multimodal-serving"]], "Multiple Profiles": [[75, "multiple-profiles"]], "NVFP4 (Blackwell)": [[85, "nvfp4-blackwell"]], "Named Arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-named-arguments"]], "Native Windows Support": [[66, "native-windows-support"]], "Natively supported models": [[15, "natively-supported-models"]], "New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget": [[22, null]], "Next Steps": [[83, "next-steps"]], "Normalization": [[78, "module-tensorrt_llm.layers.normalization"]], "Note on context outputs": [[3, "note-on-context-outputs"]], "Numerical Precision": [[85, null]], "Obtaining Arbitrary Output Tensors": [[3, "obtaining-arbitrary-output-tensors"]], "Offloading to host memory": [[8, "offloading-to-host-memory"]], "Online Serving Examples": [[58, null]], "Only collect specific iterations": [[67, "only-collect-specific-iterations"]], "OpenAI Chat Client": [[55, null], [56, null]], "OpenAI Completion Client": [[57, null]], "Option 1: Build TensorRT-LLM in One Step": [[60, "option-1-build-tensorrt-llm-in-one-step"]], "Option 1: Full Build with C++ Compilation": [[60, "option-1-full-build-with-c-compilation"]], "Option 2: Build TensorRT-LLM Step-by-Step": [[60, "option-2-build-tensorrt-llm-step-by-step"]], "Option 2: Python-Only Build without C++ Compilation": [[60, "option-2-python-only-build-without-c-compilation"]], "Other Build Modes": [[68, "other-build-modes"]], "Out of memory issues": [[18, "out-of-memory-issues"]], "Out-of-Tree Models": [[90, "out-of-tree-models"]], "Overview": [[6, "overview"], [13, "overview"], [15, "overview"], [17, "overview"], [66, null], [69, null]], "Padded and Packed Tensors": [[5, "padded-and-packed-tensors"]], "Paged Context Attention": [[75, "paged-context-attention"]], "Paged KV Cache": [[5, "paged-kv-cache"]], "Parallelism Mapping Support": [[68, "parallelism-mapping-support"]], "Parallelism Strategy": [[24, "parallelism-strategy"]], "Pattern and Pattern Manager": [[7, "pattern-and-pattern-manager"]], "Pattern-Matching and Fusion": [[14, "pattern-matching-and-fusion"]], "Performance": [[23, "performance"], [59, null], [75, "performance"]], "Performance Analysis": [[67, null]], "Performance Improvements": [[10, "performance-improvements"]], "Performance Tuning Guide": [[73, null]], "Performance expectations": [[8, "performance-expectations"]], "Performance with GEMM + SwiGLU Fusion": [[72, "performance-with-gemm-swiglu-fusion"]], "Performance with GEMM Plugin": [[75, "performance-with-gemm-plugin"]], "Performance with Low Latency GEMM plugin": [[72, "performance-with-low-latency-gemm-plugin"]], "Performance with Quantized KV Cache": [[72, "performance-with-quantized-kv-cache"]], "Performance with Reduce Norm Fusion": [[75, "performance-with-reduce-norm-fusion"]], "Performance with Reduce Norm Fusion + User Buffers:": [[72, "performance-with-reduce-norm-fusion-user-buffers"]], "Performance with multiple profiles": [[75, "performance-with-multiple-profiles"]], "Persistence mode": [[68, "persistence-mode"]], "Pipeline Parallel Reduce Scatter Optimization": [[75, "pipeline-parallel-reduce-scatter-optimization"]], "Plugin": [[80, null]], "Plugin config arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-plugin-config-arguments"]], "Plugins": [[14, "plugins"]], "Pooling": [[78, "module-tensorrt_llm.layers.pooling"]], "Postprocessing functions": [[15, "postprocessing-functions"]], "Precision Strategy": [[24, "precision-strategy"]], "Prepare": [[28, "prepare"]], "Prepare Dataset": [[70, "prepare-dataset"]], "Prepare the TensorRT-LLM Checkpoint": [[13, "prepare-the-tensorrt-llm-checkpoint"]], "Preparing a Dataset": [[68, "preparing-a-dataset"], [69, "preparing-a-dataset"]], "Prerequisite Knowledge": [[73, "prerequisite-knowledge"]], "Prerequisites": [[60, "prerequisites"], [83, "prerequisites"], [90, "prerequisites"]], "Prerequisites: Install TensorRT-LLM and download models": [[18, "prerequisites-install-tensorrt-llm-and-download-models"]], "Profiling specific iterations on a trtllm-bench/trtllm-serve run": [[67, "profiling-specific-iterations-on-a-trtllm-bench-trtllm-serve-run"]], "Prompt-Lookup-Decoding": [[10, "prompt-lookup-decoding"]], "Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs": [[24, null]], "PyExecutor": [[91, "pyexecutor"]], "PyTorch Backend": [[89, null]], "Python Bindings for the Executor API": [[3, "python-bindings-for-the-executor-api"]], "Python runtime (Not recommended to be used)": [[84, "python-runtime-not-recommended-to-be-used"]], "Quantization": [[32, "quantization"], [81, null], [89, "quantization"]], "Quantization APIs": [[17, "quantization-apis"]], "Quantization and Dequantization (Q/DQ)": [[85, "quantization-and-dequantization-q-dq"]], "Quantization in TensorRT-LLM": [[23, "quantization-in-tensorrt-llm"]], "Quantization in the PyTorch Flow": [[68, "quantization-in-the-pytorch-flow"]], "Quantized KV-Cache": [[72, "quantized-kv-cache"]], "Quick Start": [[89, "quick-start"]], "Quick Start Guide": [[83, null]], "Quickstart": [[68, "quickstart"]], "Rank Weights": [[13, "rank-weights"]], "Re-balanced the sparse experts": [[24, "re-balanced-the-sparse-experts"]], "ReDrafter": [[10, "redrafter"]], "Reduce Norm Fusion Plugin for Llama models:": [[75, "reduce-norm-fusion-plugin-for-llama-models"]], "Reduce Norm Fusion with User Buffers for Llama Models": [[72, "reduce-norm-fusion-with-user-buffers-for-llama-models"]], "Reference": [[12, "reference"], [59, null]], "Related Information": [[83, "related-information"]], "Relative Attention Bias (RAB)": [[5, "relative-attention-bias-rab"]], "Relax Acceptance Verification": [[24, "relax-acceptance-verification"]], "Release Notes": [[88, null]], "Reproducing Benchmarked Results": [[69, "reproducing-benchmarked-results"]], "Reproducing steps": [[18, "reproducing-steps"]], "Request Additional Output": [[3, "request-additional-output"]], "ResourceManager": [[91, "resourcemanager"]], "Results": [[70, "results"]], "Revisiting Paged Context Attention and Context Chunking": [[74, "revisiting-paged-context-attention-and-context-chunking"]], "Rotary Positional Embedding (RoPE)": [[5, "rotary-positional-embedding-rope"]], "RouterGEMM": [[24, "routergemm"]], "Run gpt-2b + LoRA using Executor / cpp runtime": [[9, null]], "Run the Model": [[83, "run-the-model"]], "Running Throughput and Latency Benchmarks": [[70, "running-throughput-and-latency-benchmarks"]], "Running With Weight Streaming to Reduce GPU Memory Consumption": [[11, null]], "Running multi-modal models in the PyTorch Workflow": [[68, "running-multi-modal-models-in-the-pytorch-workflow"]], "Running the Benchmark": [[69, "running-the-benchmark"]], "Running with the PyTorch Workflow": [[68, "running-with-the-pytorch-workflow"]], "Runtime": [[1, null], [14, "runtime"], [82, null]], "Runtime Customization": [[32, "runtime-customization"]], "Sampling": [[32, "sampling"]], "Sampling Parameters": [[6, "sampling-parameters"]], "Scaling factor(s)": [[5, "scaling-factor-s"]], "Scheduler": [[91, "scheduler"], [94, null]], "Scheduler Introduction": [[94, "scheduler-introduction"]], "Scripts": [[37, null], [58, null]], "Sending Requests with Different Beam Widths": [[3, "sending-requests-with-different-beam-widths"]], "Set power limits": [[68, "set-power-limits"]], "Situations that can prevent kv cache reuse": [[8, "situations-that-can-prevent-kv-cache-reuse"]], "Sliding Window Attention, Cyclic (Rolling Buffer) KV Cache": [[5, "sliding-window-attention-cyclic-rolling-buffer-kv-cache"]], "Smart Router": [[24, "smart-router"]], "Software": [[86, "software"]], "Sparse Experts as GEMMs (only works when moe_backend=CUTLASS)": [[24, "sparse-experts-as-gemms-only-works-when-moe-backend-cutlass"]], "Speculative Sampling": [[10, null]], "Speculative decoding arguments": [[25, "tensorrt_llm.commands.build-parse_arguments-speculative-decoding-arguments"]], "Speed up inference with SOTA quantization techniques in TRT-LLM": [[23, null]], "Starting a Server": [[26, "starting-a-server"]], "Step 1. Write Modeling Part": [[12, "step-1-write-modeling-part"]], "Step 2. Implement Weight Conversion": [[12, "step-2-implement-weight-conversion"]], "Step 3. Register New Model": [[12, "step-3-register-new-model"]], "Step 4. Verify New Model": [[12, "step-4-verify-new-model"]], "Step-by-Step Guide": [[90, "step-by-step-guide"]], "StreamingLLM": [[5, "streamingllm"]], "Structured output with guided decoding": [[3, "structured-output-with-guided-decoding"]], "Summary": [[68, "summary"]], "Summary of Configuration Option Recommendations:": [[72, "summary-of-configuration-option-recommendations"], [75, "summary-of-configuration-option-recommendations"]], "Support Matrix": [[86, null]], "Support matrix": [[85, "support-matrix"]], "Supported C++ Header Files": [[60, "supported-c-header-files"]], "Supported Models": [[64, "supported-models"]], "Supported Quantization Modes": [[68, "supported-quantization-modes"]], "Syntax": [[26, "syntax"]], "System Level optimizations": [[24, "system-level-optimizations"]], "TRTLLM Backend": [[24, "trtllm-backend"]], "Table of Contents": [[18, "table-of-contents"], [24, "table-of-contents"], [73, "table-of-contents"], [90, "table-of-contents"]], "Technical Detail: The QuantMode Flags": [[85, "technical-detail-the-quantmode-flags"]], "Tensor Parallel vs Expert Parallel": [[4, "tensor-parallel-vs-expert-parallel"]], "Tensor-Related Methods": [[7, "tensor-related-methods"]], "TensorRT Compiler": [[14, "tensorrt-compiler"]], "TensorRT-LLM Architecture": [[16, null]], "TensorRT-LLM Benchmarking": [[68, null]], "TensorRT-LLM Build Workflow": [[17, null]], "TensorRT-LLM Checkpoint": [[13, null]], "TensorRT-LLM Model Weights Loader": [[15, null]], "TensorRT-LLM Release 0.10.0": [[88, "tensorrt-llm-release-0-10-0"]], "TensorRT-LLM Release 0.11.0": [[88, "tensorrt-llm-release-0-11-0"]], "TensorRT-LLM Release 0.12.0": [[88, "tensorrt-llm-release-0-12-0"]], "TensorRT-LLM Release 0.13.0": [[88, "tensorrt-llm-release-0-13-0"]], "TensorRT-LLM Release 0.14.0": [[88, "tensorrt-llm-release-0-14-0"]], "TensorRT-LLM Release 0.15.0": [[88, "tensorrt-llm-release-0-15-0"]], "TensorRT-LLM Release 0.16.0": [[88, "tensorrt-llm-release-0-16-0"]], "TensorRT-LLM Release 0.17.0": [[88, "tensorrt-llm-release-0-17-0"]], "TensorRT-LLM Release 0.18.0": [[88, "tensorrt-llm-release-0-18-0"]], "TensorRT-LLM Release 0.18.1": [[88, "tensorrt-llm-release-0-18-1"]], "TensorRT-LLM Release 0.18.2": [[88, "tensorrt-llm-release-0-18-2"]], "TensorRT-LLM Release 0.19.0": [[88, "tensorrt-llm-release-0-19-0"]], "TensorRT-LLM Release 0.7.1": [[88, "tensorrt-llm-release-0-7-1"]], "TensorRT-LLM Release 0.8.0": [[88, "tensorrt-llm-release-0-8-0"]], "TensorRT-LLM Release 0.9.0": [[88, "tensorrt-llm-release-0-9-0"]], "The Executor Class": [[3, "the-executor-class"]], "The Request Class": [[3, "the-request-class"]], "The Response Class": [[3, "the-response-class"]], "The Result Class": [[3, "the-result-class"]], "Throughput Benchmarking": [[68, "throughput-benchmarking"]], "Throughput Measurements": [[69, "throughput-measurements"]], "Tips": [[87, "tips"]], "Tips and Troubleshooting": [[64, "tips-and-troubleshooting"]], "Tokenizer Customization": [[32, "tokenizer-customization"]], "Top Level API": [[91, "top-level-api"]], "Translator": [[15, "translator"]], "Trouble shooting": [[15, "trouble-shooting"]], "Troubleshooting": [[87, null]], "Troubleshooting Tips and Pitfalls To Avoid": [[70, "troubleshooting-tips-and-pitfalls-to-avoid"]], "Troubleshooting and FAQ": [[2, "troubleshooting-and-faq"]], "Tuning Case Study": [[74, "tuning-case-study"], [74, "id2"]], "Tuning Max Batch Size": [[74, "tuning-max-batch-size"]], "Tuning Max Batch Size and Max Num Tokens": [[74, null]], "Tuning Max Num Tokens": [[74, "tuning-max-num-tokens"]], "Understand inference time GPU memory usage": [[84, "understand-inference-time-gpu-memory-usage"]], "Understanding the TensorRT-LLM scheduler": [[74, "understanding-the-tensorrt-llm-scheduler"]], "Upload the Docker Image to DockerHub": [[27, "upload-the-docker-image-to-dockerhub"]], "Usage": [[2, "usage"]], "Useful Build-Time Flags": [[75, null]], "Useful Runtime Options": [[76, null]], "Using Medusa with TensorRT-LLM": [[10, "using-medusa-with-tensorrt-llm"]], "Validated Networks for Benchmarking": [[68, "validated-networks-for-benchmarking"]], "Variables": [[69, "variables"]], "Visualize the PyTorch profiler results": [[67, "visualize-the-pytorch-profiler-results"]], "WIP: Chunked context support on DeepSeek models": [[18, "wip-chunked-context-support-on-deepseek-models"]], "WIP: Enable more features by default": [[18, "wip-enable-more-features-by-default"]], "Weight Bindings": [[14, "weight-bindings"]], "Weight Loading": [[90, "weight-loading"]], "Welcome to TensorRT-LLM\u2019s Documentation!": [[59, null]], "What Can You Do With TensorRT-LLM?": [[66, "what-can-you-do-with-tensorrt-llm"]], "What is H100 FP8?": [[20, "what-is-h100-fp8"]], "What\u2019s coming next": [[23, "whats-coming-next"]], "When to Use Graph Rewriting?": [[7, "when-to-use-graph-rewriting"]], "Workflow": [[15, "workflow"], [68, "workflow"]], "Workload Profile": [[24, "workload-profile"]], "World Configuration": [[6, "world-configuration"]], "XQA Optimization": [[5, "xqa-optimization"]], "bufferManager.h": [[1, "buffermanager-h"]], "cacheCommunicator.h": [[0, "cachecommunicator-h"]], "common.h": [[1, "common-h"]], "cudaEvent.h": [[1, "cudaevent-h"]], "cudaStream.h": [[1, "cudastream-h"]], "dataTransceiverState.h": [[0, "datatransceiverstate-h"]], "decoderState.h": [[1, "decoderstate-h"]], "decodingInput.h": [[1, "decodinginput-h"]], "decodingOutput.h": [[1, "decodingoutput-h"]], "disaggServerUtil.h": [[0, "disaggserverutil-h"]], "disaggregated": [[26, "trtllm-serve-disaggregated"]], "disaggregated_mpi_worker": [[26, "trtllm-serve-disaggregated-mpi-worker"]], "eagleBuffers.h": [[1, "eaglebuffers-h"]], "eagleModule.h": [[1, "eaglemodule-h"]], "executor.h": [[0, "executor-h"]], "explicitDraftTokensBuffers.h": [[1, "explicitdrafttokensbuffers-h"]], "gptDecoder.h": [[1, "gptdecoder-h"]], "gptDecoderBatched.h": [[1, "gptdecoderbatched-h"]], "gptJsonConfig.h": [[1, "gptjsonconfig-h"]], "iBuffer.h": [[1, "ibuffer-h"]], "iGptDecoderBatched.h": [[1, "igptdecoderbatched-h"]], "iTensor.h": [[1, "itensor-h"]], "ipcNvlsMemory.h": [[1, "ipcnvlsmemory-h"]], "ipcUtils.h": [[1, "ipcutils-h"]], "lookaheadBuffers.h": [[1, "lookaheadbuffers-h"]], "lookaheadModule.h": [[1, "lookaheadmodule-h"]], "loraCache.h": [[1, "loracache-h"]], "loraCachePageManagerConfig.h": [[1, "loracachepagemanagerconfig-h"]], "loraModule.h": [[1, "loramodule-h"]], "medusaModule.h": [[1, "medusamodule-h"]], "memoryCounters.h": [[1, "memorycounters-h"]], "modelConfig.h": [[1, "modelconfig-h"]], "promptTuningParams.h": [[1, "prompttuningparams-h"]], "rawEngine.h": [[1, "rawengine-h"]], "request.h": [[1, "request-h"]], "runtimeDefaults.h": [[1, "runtimedefaults-h"]], "samplingConfig.h": [[1, "samplingconfig-h"]], "serialization.h": [[0, "serialization-h"]], "serve": [[26, "trtllm-serve-serve"]], "speculativeDecodingMode.h": [[1, "speculativedecodingmode-h"]], "speculativeDecodingModule.h": [[1, "speculativedecodingmodule-h"]], "tensor.h": [[0, "tensor-h"]], "tllmLogger.h": [[1, "tllmlogger-h"]], "trtllm-build": [[25, null]], "trtllm-serve": [[26, null], [26, "trtllm-serve"]], "types.h": [[0, "types-h"]], "worldConfig.h": [[1, "worldconfig-h"]]}, "docnames": ["_cpp_gen/executor", "_cpp_gen/runtime", "advanced/disaggregated-service", "advanced/executor", "advanced/expert-parallelism", "advanced/gpt-attention", "advanced/gpt-runtime", "advanced/graph-rewriting", "advanced/kv-cache-reuse", "advanced/lora", "advanced/speculative-decoding", "advanced/weight-streaming", "architecture/add-model", "architecture/checkpoint", "architecture/core-concepts", "architecture/model-weights-loader", "architecture/overview", "architecture/workflow", "blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM", "blogs/Falcon180B-H200", "blogs/H100vsA100", "blogs/H200launch", "blogs/XQA-kernel", "blogs/quantization-in-TRT-LLM", "blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs", "commands/trtllm-build", "commands/trtllm-serve", "dev-on-cloud/build-image-to-dockerhub", "dev-on-cloud/dev-on-runpod", "examples/curl_chat_client", "examples/curl_chat_client_for_multimodal", "examples/curl_completion_client", "examples/customization", "examples/deepseek_r1_reasoning_parser", "examples/genai_perf_client", "examples/genai_perf_client_for_multimodal", "examples/index", "examples/llm_api_examples", "examples/llm_auto_parallel", "examples/llm_eagle_decoding", "examples/llm_guided_decoding", "examples/llm_inference", "examples/llm_inference_async", "examples/llm_inference_async_streaming", "examples/llm_inference_customize", "examples/llm_inference_distributed", "examples/llm_inference_kv_events", "examples/llm_logits_processor", "examples/llm_lookahead_decoding", "examples/llm_medusa_decoding", "examples/llm_mgmn_llm_distributed", "examples/llm_mgmn_trtllm_bench", "examples/llm_mgmn_trtllm_serve", "examples/llm_multilora", "examples/llm_quantization", "examples/openai_chat_client", "examples/openai_chat_client_for_multimodal", "examples/openai_completion_client", "examples/trtllm_serve_examples", "index", "installation/build-from-source-linux", "installation/grace-hopper", "installation/linux", "key-features", "llm-api/index", "llm-api/reference", "overview", "performance/perf-analysis", "performance/perf-benchmarking", "performance/perf-overview", "performance/performance-tuning-guide/benchmarking-default-performance", "performance/performance-tuning-guide/deciding-model-sharding-strategy", "performance/performance-tuning-guide/fp8-quantization", "performance/performance-tuning-guide/index", "performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens", "performance/performance-tuning-guide/useful-build-time-flags", "performance/performance-tuning-guide/useful-runtime-flags", "python-api/tensorrt_llm.functional", "python-api/tensorrt_llm.layers", "python-api/tensorrt_llm.models", "python-api/tensorrt_llm.plugin", "python-api/tensorrt_llm.quantization", "python-api/tensorrt_llm.runtime", "quick-start-guide", "reference/memory", "reference/precision", "reference/support-matrix", "reference/troubleshooting", "release-notes", "torch", "torch/adding_new_model", "torch/arch_overview", "torch/attention", "torch/kv_cache_manager", "torch/scheduler"], "envversion": {"sphinx": 62, "sphinx.domains.c": 3, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 9, "sphinx.domains.index": 1, "sphinx.domains.javascript": 3, "sphinx.domains.math": 2, "sphinx.domains.python": 4, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.todo": 2, "sphinx.ext.viewcode": 1}, "filenames": ["_cpp_gen/executor.rst", "_cpp_gen/runtime.rst", "advanced/disaggregated-service.md", "advanced/executor.md", "advanced/expert-parallelism.md", "advanced/gpt-attention.md", "advanced/gpt-runtime.md", "advanced/graph-rewriting.md", "advanced/kv-cache-reuse.md", "advanced/lora.md", "advanced/speculative-decoding.md", "advanced/weight-streaming.md", "architecture/add-model.md", "architecture/checkpoint.md", "architecture/core-concepts.md", "architecture/model-weights-loader.md", "architecture/overview.md", "architecture/workflow.md", "blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md", "blogs/Falcon180B-H200.md", "blogs/H100vsA100.md", "blogs/H200launch.md", "blogs/XQA-kernel.md", "blogs/quantization-in-TRT-LLM.md", "blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs.md", "commands/trtllm-build.rst", "commands/trtllm-serve.rst", "dev-on-cloud/build-image-to-dockerhub.md", "dev-on-cloud/dev-on-runpod.md", "examples/curl_chat_client.rst", "examples/curl_chat_client_for_multimodal.rst", "examples/curl_completion_client.rst", "examples/customization.md", "examples/deepseek_r1_reasoning_parser.rst", "examples/genai_perf_client.rst", "examples/genai_perf_client_for_multimodal.rst", "examples/index.rst", "examples/llm_api_examples.rst", "examples/llm_auto_parallel.rst", "examples/llm_eagle_decoding.rst", "examples/llm_guided_decoding.rst", "examples/llm_inference.rst", "examples/llm_inference_async.rst", "examples/llm_inference_async_streaming.rst", "examples/llm_inference_customize.rst", "examples/llm_inference_distributed.rst", "examples/llm_inference_kv_events.rst", "examples/llm_logits_processor.rst", "examples/llm_lookahead_decoding.rst", "examples/llm_medusa_decoding.rst", "examples/llm_mgmn_llm_distributed.rst", "examples/llm_mgmn_trtllm_bench.rst", "examples/llm_mgmn_trtllm_serve.rst", "examples/llm_multilora.rst", "examples/llm_quantization.rst", "examples/openai_chat_client.rst", "examples/openai_chat_client_for_multimodal.rst", "examples/openai_completion_client.rst", "examples/trtllm_serve_examples.rst", "index.rst", "installation/build-from-source-linux.md", "installation/grace-hopper.md", "installation/linux.md", "key-features.md", "llm-api/index.md", "llm-api/reference.rst", "overview.md", "performance/perf-analysis.md", "performance/perf-benchmarking.md", "performance/perf-overview.md", "performance/performance-tuning-guide/benchmarking-default-performance.md", "performance/performance-tuning-guide/deciding-model-sharding-strategy.md", "performance/performance-tuning-guide/fp8-quantization.md", "performance/performance-tuning-guide/index.rst", "performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.md", "performance/performance-tuning-guide/useful-build-time-flags.md", "performance/performance-tuning-guide/useful-runtime-flags.md", "python-api/tensorrt_llm.functional.rst", "python-api/tensorrt_llm.layers.rst", "python-api/tensorrt_llm.models.rst", "python-api/tensorrt_llm.plugin.rst", "python-api/tensorrt_llm.quantization.rst", "python-api/tensorrt_llm.runtime.rst", "quick-start-guide.md", "reference/memory.md", "reference/precision.md", "reference/support-matrix.md", "reference/troubleshooting.md", "release-notes.md", "torch.md", "torch/adding_new_model.md", "torch/arch_overview.md", "torch/attention.md", "torch/kv_cache_manager.md", "torch/scheduler.md"], "indexentries": {"--backend": [[26, "cmdoption-trtllm-serve-serve-backend", false]], "--cluster_size": [[26, "cmdoption-trtllm-serve-serve-cluster_size", false]], "--config_file": [[26, "cmdoption-trtllm-serve-disaggregated-c", false], [26, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false]], "--ep_size": [[26, "cmdoption-trtllm-serve-serve-ep_size", false]], "--extra_llm_api_options": [[26, "cmdoption-trtllm-serve-serve-extra_llm_api_options", false]], "--gpus_per_node": [[26, "cmdoption-trtllm-serve-serve-gpus_per_node", false]], "--host": [[26, "cmdoption-trtllm-serve-serve-host", false]], "--kv_cache_free_gpu_memory_fraction": [[26, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", false]], "--log_level": [[26, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", false], [26, "cmdoption-trtllm-serve-serve-log_level", false]], "--max_batch_size": [[26, "cmdoption-trtllm-serve-serve-max_batch_size", false]], "--max_beam_width": [[26, "cmdoption-trtllm-serve-serve-max_beam_width", false]], "--max_num_tokens": [[26, "cmdoption-trtllm-serve-serve-max_num_tokens", false]], "--max_seq_len": [[26, "cmdoption-trtllm-serve-serve-max_seq_len", false]], "--num_postprocess_workers": [[26, "cmdoption-trtllm-serve-serve-num_postprocess_workers", false]], "--port": [[26, "cmdoption-trtllm-serve-serve-port", false]], "--pp_size": [[26, "cmdoption-trtllm-serve-serve-pp_size", false]], "--reasoning_parser": [[26, "cmdoption-trtllm-serve-serve-reasoning_parser", false]], "--request_timeout": [[26, "cmdoption-trtllm-serve-disaggregated-r", false]], "--server_start_timeout": [[26, "cmdoption-trtllm-serve-disaggregated-t", false]], "--tokenizer": [[26, "cmdoption-trtllm-serve-serve-tokenizer", false]], "--tp_size": [[26, "cmdoption-trtllm-serve-serve-tp_size", false]], "--trust_remote_code": [[26, "cmdoption-trtllm-serve-serve-trust_remote_code", false]], "-c": [[26, "cmdoption-trtllm-serve-disaggregated-c", false], [26, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false]], "-r": [[26, "cmdoption-trtllm-serve-disaggregated-r", false]], "-t": [[26, "cmdoption-trtllm-serve-disaggregated-t", false]], "__init__() (tensorrt_llm.llmapi.buildcacheconfig method)": [[65, "tensorrt_llm.llmapi.BuildCacheConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.buildconfig method)": [[65, "tensorrt_llm.llmapi.BuildConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.completionoutput method)": [[65, "tensorrt_llm.llmapi.CompletionOutput.__init__", false]], "__init__() (tensorrt_llm.llmapi.disaggregatedparams method)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.__init__", false]], "__init__() (tensorrt_llm.llmapi.guideddecodingparams method)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.__init__", false]], "__init__() (tensorrt_llm.llmapi.kvcacheretentionconfig method)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig method)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.__init__", false]], "__init__() (tensorrt_llm.llmapi.lookaheaddecodingconfig method)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.__init__", false]], "__init__() (tensorrt_llm.llmapi.quantconfig method)": [[65, "tensorrt_llm.llmapi.QuantConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.requestoutput method)": [[65, "tensorrt_llm.llmapi.RequestOutput.__init__", false]], "__init__() (tensorrt_llm.llmapi.samplingparams method)": [[65, "tensorrt_llm.llmapi.SamplingParams.__init__", false]], "abort() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.abort", false]], "abs() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.abs", false]], "abs() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.abs", false]], "activation() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.activation", false]], "adalayernorm (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNorm", false]], "adalayernormcontinuous (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormContinuous", false]], "adalayernormzero (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormZero", false]], "adalayernormzerosingle (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle", false]], "add() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.add", false]], "add_input() (tensorrt_llm.functional.conditional method)": [[77, "tensorrt_llm.functional.Conditional.add_input", false]], "add_output() (tensorrt_llm.functional.conditional method)": [[77, "tensorrt_llm.functional.Conditional.add_output", false]], "add_sequence() (tensorrt_llm.runtime.kvcachemanager method)": [[82, "tensorrt_llm.runtime.KVCacheManager.add_sequence", false]], "add_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.add_special_tokens", false]], "additional_model_outputs (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.additional_model_outputs", false]], "alibi (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.alibi", false]], "alibi_with_scale (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.alibi_with_scale", false]], "allgather() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.allgather", false]], "allreduce() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.allreduce", false]], "allreducefusionop (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.AllReduceFusionOp", false]], "allreduceparams (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.AllReduceParams", false]], "allreducestrategy (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.AllReduceStrategy", false]], "apply_batched_logits_processor (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.apply_batched_logits_processor", false]], "apply_llama3_scaling() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_llama3_scaling", false]], "apply_rotary_pos_emb() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb", false]], "apply_rotary_pos_emb_chatglm() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb_chatglm", false]], "apply_rotary_pos_emb_cogvlm() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb_cogvlm", false]], "arange() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.arange", false]], "argmax() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.argmax", false]], "assert_valid_quant_algo() (tensorrt_llm.models.gemmaforcausallm class method)": [[79, "tensorrt_llm.models.GemmaForCausalLM.assert_valid_quant_algo", false]], "assertion() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.assertion", false]], "attention (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.Attention", false]], "attentionmaskparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.AttentionMaskParams", false]], "attentionmasktype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.AttentionMaskType", false]], "attentionparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.AttentionParams", false]], "attn_processors (tensorrt_llm.models.sd3transformer2dmodel property)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.attn_processors", false]], "audio_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.audio_engine_dir", false]], "auto (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.AUTO", false]], "auto_parallel_config (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.auto_parallel_config", false]], "avg_pool2d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.avg_pool2d", false]], "avgpool2d (class in tensorrt_llm.layers.pooling)": [[78, "tensorrt_llm.layers.pooling.AvgPool2d", false]], "axes (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.axes", false]], "bad (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.bad", false]], "bad_token_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.bad_token_ids", false]], "bad_words_list (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.bad_words_list", false]], "baichuanforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BaichuanForCausalLM", false]], "batch_size (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.batch_size", false]], "batchingtype (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.BatchingType", false]], "beam_search_diversity_rate (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.beam_search_diversity_rate", false]], "beam_search_diversity_rate (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.beam_search_diversity_rate", false]], "beam_width_array (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.beam_width_array", false]], "bert_attention() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.bert_attention", false]], "bertattention (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.BertAttention", false]], "bertforquestionanswering (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BertForQuestionAnswering", false]], "bertforsequenceclassification (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BertForSequenceClassification", false]], "bertmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BertModel", false]], "best_of (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.best_of", false]], "bidirectional (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.bidirectional", false]], "bidirectionalglm (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.bidirectionalglm", false]], "blocksparse (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.blocksparse", false]], "blocksparseattnparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.BlockSparseAttnParams", false]], "bloomforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BloomForCausalLM", false]], "bloommodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.BloomModel", false]], "broadcast_helper() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.broadcast_helper", false]], "buffer_allocated (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.buffer_allocated", false]], "buildcacheconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.BuildCacheConfig", false]], "buildconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.BuildConfig", false]], "cache_root (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildCacheConfig.cache_root", false]], "cache_root (tensorrt_llm.llmapi.buildcacheconfig property)": [[65, "id7", false]], "cachetransceiverconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.CacheTransceiverConfig", false]], "calculate_speculative_resource() (tensorrt_llm.llmapi.lookaheaddecodingconfig method)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.calculate_speculative_resource", false]], "calib_batch_size (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.calib_batch_size", false]], "calib_batches (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.calib_batches", false]], "calib_dataset (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.calib_dataset", false]], "calib_max_seq_length (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.calib_max_seq_length", false]], "calibconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.CalibConfig", false]], "capacity_scheduler_policy (tensorrt_llm.llmapi.schedulerconfig attribute)": [[65, "tensorrt_llm.llmapi.SchedulerConfig.capacity_scheduler_policy", false]], "capacityschedulerpolicy (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.CapacitySchedulerPolicy", false]], "cast (class in tensorrt_llm.layers.cast)": [[78, "tensorrt_llm.layers.cast.Cast", false]], "cast() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.cast", false]], "cast() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.cast", false]], "categorical_sample() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.categorical_sample", false]], "causal (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.causal", false]], "chatglm (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.chatglm", false]], "chatglmconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.ChatGLMConfig", false]], "chatglmforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.ChatGLMForCausalLM", false]], "chatglmgenerationsession (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.ChatGLMGenerationSession", false]], "chatglmmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.ChatGLMModel", false]], "check_config() (tensorrt_llm.models.decodermodel method)": [[79, "tensorrt_llm.models.DecoderModel.check_config", false]], "check_config() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.check_config", false]], "check_config() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.check_config", false]], "check_config() (tensorrt_llm.models.falconforcausallm method)": [[79, "tensorrt_llm.models.FalconForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.mptforcausallm method)": [[79, "tensorrt_llm.models.MPTForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.optforcausallm method)": [[79, "tensorrt_llm.models.OPTForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.phiforcausallm method)": [[79, "tensorrt_llm.models.PhiForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.pretrainedmodel method)": [[79, "tensorrt_llm.models.PretrainedModel.check_config", false]], "choices() (tensorrt_llm.functional.positionembeddingtype static method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.choices", false]], "chunk() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.chunk", false]], "clamp_val (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.clamp_val", false]], "clip() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.clip", false]], "clipvisiontransformer (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.CLIPVisionTransformer", false]], "cogvlmattention (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.CogVLMAttention", false]], "cogvlmconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.CogVLMConfig", false]], "cogvlmforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.CogVLMForCausalLM", false]], "cohereforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.CohereForCausalLM", false]], "collect_and_bias() (tensorrt_llm.layers.linear.linear method)": [[78, "tensorrt_llm.layers.linear.Linear.collect_and_bias", false]], "collect_and_bias() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.collect_and_bias", false]], "collect_and_bias() (tensorrt_llm.layers.linear.rowlinear method)": [[78, "tensorrt_llm.layers.linear.RowLinear.collect_and_bias", false]], "columnlinear (in module tensorrt_llm.layers.linear)": [[78, "tensorrt_llm.layers.linear.ColumnLinear", false]], "combinedtimesteplabelembeddings (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings", false]], "combinedtimesteptextprojembeddings (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings", false]], "completionoutput (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.CompletionOutput", false]], "compute_relative_bias() (in module tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.compute_relative_bias", false]], "concat() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.concat", false]], "conditional (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.Conditional", false]], "config_class (tensorrt_llm.models.baichuanforcausallm attribute)": [[79, "tensorrt_llm.models.BaichuanForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.chatglmforcausallm attribute)": [[79, "tensorrt_llm.models.ChatGLMForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.cogvlmforcausallm attribute)": [[79, "tensorrt_llm.models.CogVLMForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.cohereforcausallm attribute)": [[79, "tensorrt_llm.models.CohereForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.dbrxforcausallm attribute)": [[79, "tensorrt_llm.models.DbrxForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.deepseekforcausallm attribute)": [[79, "tensorrt_llm.models.DeepseekForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.deepseekv2forcausallm attribute)": [[79, "tensorrt_llm.models.DeepseekV2ForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.eagleforcausallm attribute)": [[79, "tensorrt_llm.models.EagleForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.falconforcausallm attribute)": [[79, "tensorrt_llm.models.FalconForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gemmaforcausallm attribute)": [[79, "tensorrt_llm.models.GemmaForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gptforcausallm attribute)": [[79, "tensorrt_llm.models.GPTForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gptjforcausallm attribute)": [[79, "tensorrt_llm.models.GPTJForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.llamaforcausallm attribute)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.mambaforcausallm attribute)": [[79, "tensorrt_llm.models.MambaForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.medusaforcausallm attribute)": [[79, "tensorrt_llm.models.MedusaForCausalLm.config_class", false]], "config_class (tensorrt_llm.models.mllamaforcausallm attribute)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.phi3forcausallm attribute)": [[79, "tensorrt_llm.models.Phi3ForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.phiforcausallm attribute)": [[79, "tensorrt_llm.models.PhiForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.sd3transformer2dmodel attribute)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.config_class", false]], "constant() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.constant", false]], "constant_to_tensor_() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.constant_to_tensor_", false]], "constants_to_tensors_() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.constants_to_tensors_", false]], "context (tensorrt_llm.runtime.session property)": [[82, "tensorrt_llm.runtime.Session.context", false]], "context_chunking_policy (tensorrt_llm.llmapi.schedulerconfig attribute)": [[65, "tensorrt_llm.llmapi.SchedulerConfig.context_chunking_policy", false]], "context_logits (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.context_logits", false]], "context_mem_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.context_mem_size", false]], "context_mem_size (tensorrt_llm.runtime.session property)": [[82, "tensorrt_llm.runtime.Session.context_mem_size", false]], "contextchunkingpolicy (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.ContextChunkingPolicy", false]], "conv1d (class in tensorrt_llm.layers.conv)": [[78, "tensorrt_llm.layers.conv.Conv1d", false]], "conv1d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.conv1d", false]], "conv2d (class in tensorrt_llm.layers.conv)": [[78, "tensorrt_llm.layers.conv.Conv2d", false]], "conv2d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.conv2d", false]], "conv3d (class in tensorrt_llm.layers.conv)": [[78, "tensorrt_llm.layers.conv.Conv3d", false]], "conv3d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.conv3d", false]], "conv_kernel (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.conv_kernel", false]], "conv_kernel (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.conv_kernel", false]], "conv_transpose2d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.conv_transpose2d", false]], "convtranspose2d (class in tensorrt_llm.layers.conv)": [[78, "tensorrt_llm.layers.conv.ConvTranspose2d", false]], "copy_on_partial_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.copy_on_partial_reuse", false]], "cos() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.cos", false]], "cp_split_plugin() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.cp_split_plugin", false]], "cpp_e2e (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.cpp_e2e", false]], "cpp_llm_only (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.cpp_llm_only", false]], "create_allreduce_plugin() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.create_allreduce_plugin", false]], "create_attention_const_params() (tensorrt_llm.layers.attention.attention static method)": [[78, "tensorrt_llm.layers.attention.Attention.create_attention_const_params", false]], "create_fake_weight() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_fake_weight", false]], "create_runtime_defaults() (tensorrt_llm.models.pretrainedconfig static method)": [[79, "tensorrt_llm.models.PretrainedConfig.create_runtime_defaults", false]], "create_sinusoidal_positions() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions", false]], "create_sinusoidal_positions_for_attention_plugin() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_for_attention_plugin", false]], "create_sinusoidal_positions_for_cogvlm_attention_plugin() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_for_cogvlm_attention_plugin", false]], "create_sinusoidal_positions_long_rope() (tensorrt_llm.functional.ropeembeddingutils method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_long_rope", false]], "create_sinusoidal_positions_yarn() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_yarn", false]], "cropped_pos_embed() (tensorrt_llm.layers.embedding.sd3patchembed method)": [[78, "tensorrt_llm.layers.embedding.SD3PatchEmbed.cropped_pos_embed", false]], "cross_attention (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.cross_attention", false]], "cross_attention (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.cross_attention", false]], "cross_kv_cache_fraction (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.cross_kv_cache_fraction", false]], "ctx_request_id (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.ctx_request_id", false]], "cuda_graph_cache_size (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.cuda_graph_cache_size", false]], "cuda_graph_mode (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.cuda_graph_mode", false]], "cuda_graph_mode (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.cuda_graph_mode", false]], "cuda_stream_guard() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.cuda_stream_guard", false]], "cuda_stream_sync() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.cuda_stream_sync", false]], "cumsum() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.cumsum", false]], "cumulative_logprob (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.cumulative_logprob", false]], "custom_mask (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.custom_mask", false]], "data (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.data", false]], "dbrxconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DbrxConfig", false]], "dbrxforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DbrxForCausalLM", false]], "debug_mode (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.debug_mode", false]], "debug_tensors_to_save (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.debug_tensors_to_save", false]], "decode() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.decode", false]], "decode_batch() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.decode_batch", false]], "decode_duration_ms (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.decode_duration_ms", false]], "decode_regular() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.decode_regular", false]], "decode_retention_priority (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.decode_retention_priority", false]], "decode_stream() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.decode_stream", false]], "decode_words_list() (in module tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.decode_words_list", false]], "decodermodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DecoderModel", false]], "decoding_type (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.decoding_type", false]], "deepseekforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DeepseekForCausalLM", false]], "deepseekv2attention (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.DeepseekV2Attention", false]], "deepseekv2forcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DeepseekV2ForCausalLM", false]], "default_plugin_config() (tensorrt_llm.models.cogvlmforcausallm method)": [[79, "tensorrt_llm.models.CogVLMForCausalLM.default_plugin_config", false]], "default_plugin_config() (tensorrt_llm.models.llamaforcausallm method)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.default_plugin_config", false]], "deferred (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.deferred", false]], "detokenize (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.detokenize", false]], "device (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.device", false]], "device (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.device", false]], "diffusersattention (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.DiffusersAttention", false]], "dimrange (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.DimRange", false]], "disable (tensorrt_llm.functional.sidestreamidtype attribute)": [[77, "tensorrt_llm.functional.SideStreamIDType.disable", false]], "disable_forward_chunking() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.disable_forward_chunking", false]], "disaggregated_params (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.disaggregated_params", false]], "disaggregatedparams (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams", false]], "dit (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.DiT", false]], "div() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.div", false]], "dora_plugin() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.dora_plugin", false]], "draft_tokens (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.draft_tokens", false]], "draft_tokens_external (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.DRAFT_TOKENS_EXTERNAL", false]], "dry_run (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.dry_run", false]], "dtype (tensorrt_llm.functional.tensor property)": [[77, "tensorrt_llm.functional.Tensor.dtype", false]], "dtype (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.dtype", false]], "dtype (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.dtype", false]], "dtype (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.dtype", false]], "dtype (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.dtype", false]], "dtype (tensorrt_llm.runtime.tensorinfo attribute)": [[82, "tensorrt_llm.runtime.TensorInfo.dtype", false]], "dump_debug_buffers() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.dump_debug_buffers", false]], "duration_ms (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.duration_ms", false]], "dynamic (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.dynamic", false]], "dynamic_batch_config (tensorrt_llm.llmapi.schedulerconfig attribute)": [[65, "tensorrt_llm.llmapi.SchedulerConfig.dynamic_batch_config", false]], "dynamic_batch_moving_average_window (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[65, "tensorrt_llm.llmapi.DynamicBatchConfig.dynamic_batch_moving_average_window", false]], "dynamic_tree_max_topk (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.dynamic_tree_max_topK", false]], "dynamicbatchconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.DynamicBatchConfig", false]], "eagle (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.EAGLE", false]], "eagle_choices (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.eagle_choices", false]], "eagledecodingconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig", false]], "eagleforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.EagleForCausalLM", false]], "early_stop_criteria() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.early_stop_criteria", false]], "early_stopping (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.early_stopping", false]], "early_stopping (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.early_stopping", false]], "einsum() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.einsum", false]], "elementwise_binary() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.elementwise_binary", false]], "embedding (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.Embedding", false]], "embedding() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.embedding", false]], "embedding_bias (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.embedding_bias", false]], "enable_batch_size_tuning (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[65, "tensorrt_llm.llmapi.DynamicBatchConfig.enable_batch_size_tuning", false]], "enable_block_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.enable_block_reuse", false]], "enable_context_fmha_fp32_acc (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.enable_context_fmha_fp32_acc", false]], "enable_debug_output (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.enable_debug_output", false]], "enable_forward_chunking() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.enable_forward_chunking", false]], "enable_max_num_tokens_tuning (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[65, "tensorrt_llm.llmapi.DynamicBatchConfig.enable_max_num_tokens_tuning", false]], "enable_partial_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.enable_partial_reuse", false]], "encdecmodelrunner (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.EncDecModelRunner", false]], "encoder_run() (tensorrt_llm.runtime.encdecmodelrunner method)": [[82, "tensorrt_llm.runtime.EncDecModelRunner.encoder_run", false]], "encodermodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.EncoderModel", false]], "end_id (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.end_id", false]], "end_id (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.end_id", false]], "engine (tensorrt_llm.runtime.session property)": [[82, "tensorrt_llm.runtime.Session.engine", false]], "engine_inspector (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.engine_inspector", false]], "eq() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.eq", false]], "equal_progress (tensorrt_llm.llmapi.contextchunkingpolicy attribute)": [[65, "tensorrt_llm.llmapi.ContextChunkingPolicy.EQUAL_PROGRESS", false]], "event_buffer_max_size (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.event_buffer_max_size", false]], "exclude_input_from_output (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.exclude_input_from_output", false]], "exclude_modules (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.exclude_modules", false]], "exp() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.exp", false]], "expand() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.expand", false]], "expand_dims() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.expand_dims", false]], "expand_dims_like() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.expand_dims_like", false]], "expand_mask() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.expand_mask", false]], "explicit_draft_tokens (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.EXPLICIT_DRAFT_TOKENS", false]], "extendedruntimeperfknobconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig", false]], "falconconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.FalconConfig", false]], "falconforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.FalconForCausalLM", false]], "falconmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.FalconModel", false]], "fc_gate() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[78, "tensorrt_llm.layers.mlp.FusedGatedMLP.fc_gate", false]], "fc_gate_dora() (in module tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.fc_gate_dora", false]], "fc_gate_lora() (in module tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.fc_gate_lora", false]], "fc_gate_plugin() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[78, "tensorrt_llm.layers.mlp.FusedGatedMLP.fc_gate_plugin", false]], "fill_attention_const_params_for_long_rope() (tensorrt_llm.layers.attention.attentionparams method)": [[78, "tensorrt_llm.layers.attention.AttentionParams.fill_attention_const_params_for_long_rope", false]], "fill_attention_const_params_for_rope() (tensorrt_llm.layers.attention.attentionparams method)": [[78, "tensorrt_llm.layers.attention.AttentionParams.fill_attention_const_params_for_rope", false]], "fill_attention_params() (tensorrt_llm.layers.attention.attention static method)": [[78, "tensorrt_llm.layers.attention.Attention.fill_attention_params", false]], "fill_none_tensor_list() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[78, "tensorrt_llm.layers.attention.KeyValueCacheParams.fill_none_tensor_list", false]], "fill_value (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.fill_value", false]], "filter_medusa_logits() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.filter_medusa_logits", false]], "finalize_decoder() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.finalize_decoder", false]], "find_best_medusa_path() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.find_best_medusa_path", false]], "finish_reason (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.finish_reason", false]], "finished (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.finished", false]], "first_come_first_served (tensorrt_llm.llmapi.contextchunkingpolicy attribute)": [[65, "tensorrt_llm.llmapi.ContextChunkingPolicy.FIRST_COME_FIRST_SERVED", false]], "first_gen_tokens (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.first_gen_tokens", false]], "first_layer (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.first_layer", false]], "flatten() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.flatten", false]], "flatten() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.flatten", false]], "flip() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.flip", false]], "floordiv() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.floordiv", false]], "fmt_dim (c macro)": [[1, "c.FMT_DIM", false]], "for_each_rank() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.for_each_rank", false]], "force_num_profiles (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.force_num_profiles", false]], "forward() (tensorrt_llm.layers.activation.mish method)": [[78, "tensorrt_llm.layers.activation.Mish.forward", false]], "forward() (tensorrt_llm.layers.attention.attention method)": [[78, "tensorrt_llm.layers.attention.Attention.forward", false]], "forward() (tensorrt_llm.layers.attention.bertattention method)": [[78, "tensorrt_llm.layers.attention.BertAttention.forward", false]], "forward() (tensorrt_llm.layers.attention.cogvlmattention method)": [[78, "tensorrt_llm.layers.attention.CogVLMAttention.forward", false]], "forward() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[78, "tensorrt_llm.layers.attention.DeepseekV2Attention.forward", false]], "forward() (tensorrt_llm.layers.attention.diffusersattention method)": [[78, "tensorrt_llm.layers.attention.DiffusersAttention.forward", false]], "forward() (tensorrt_llm.layers.cast.cast method)": [[78, "tensorrt_llm.layers.cast.Cast.forward", false]], "forward() (tensorrt_llm.layers.conv.conv1d method)": [[78, "tensorrt_llm.layers.conv.Conv1d.forward", false]], "forward() (tensorrt_llm.layers.conv.conv2d method)": [[78, "tensorrt_llm.layers.conv.Conv2d.forward", false]], "forward() (tensorrt_llm.layers.conv.conv3d method)": [[78, "tensorrt_llm.layers.conv.Conv3d.forward", false]], "forward() (tensorrt_llm.layers.conv.convtranspose2d method)": [[78, "tensorrt_llm.layers.conv.ConvTranspose2d.forward", false]], "forward() (tensorrt_llm.layers.embedding.combinedtimesteplabelembeddings method)": [[78, "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings.forward", false]], "forward() (tensorrt_llm.layers.embedding.combinedtimesteptextprojembeddings method)": [[78, "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings.forward", false]], "forward() (tensorrt_llm.layers.embedding.embedding method)": [[78, "tensorrt_llm.layers.embedding.Embedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.labelembedding method)": [[78, "tensorrt_llm.layers.embedding.LabelEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.pixartalphatextprojection method)": [[78, "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection.forward", false]], "forward() (tensorrt_llm.layers.embedding.prompttuningembedding method)": [[78, "tensorrt_llm.layers.embedding.PromptTuningEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.sd3patchembed method)": [[78, "tensorrt_llm.layers.embedding.SD3PatchEmbed.forward", false]], "forward() (tensorrt_llm.layers.embedding.timestepembedding method)": [[78, "tensorrt_llm.layers.embedding.TimestepEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.timesteps method)": [[78, "tensorrt_llm.layers.embedding.Timesteps.forward", false]], "forward() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.forward", false]], "forward() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[78, "tensorrt_llm.layers.mlp.FusedGatedMLP.forward", false]], "forward() (tensorrt_llm.layers.mlp.gatedmlp method)": [[78, "tensorrt_llm.layers.mlp.GatedMLP.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearactivation method)": [[78, "tensorrt_llm.layers.mlp.LinearActivation.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearapproximategelu method)": [[78, "tensorrt_llm.layers.mlp.LinearApproximateGELU.forward", false]], "forward() (tensorrt_llm.layers.mlp.lineargeglu method)": [[78, "tensorrt_llm.layers.mlp.LinearGEGLU.forward", false]], "forward() (tensorrt_llm.layers.mlp.lineargelu method)": [[78, "tensorrt_llm.layers.mlp.LinearGELU.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearswiglu method)": [[78, "tensorrt_llm.layers.mlp.LinearSwiGLU.forward", false]], "forward() (tensorrt_llm.layers.mlp.mlp method)": [[78, "tensorrt_llm.layers.mlp.MLP.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernorm method)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormcontinuous method)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormContinuous.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormzero method)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormZero.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormzerosingle method)": [[78, "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle.forward", false]], "forward() (tensorrt_llm.layers.normalization.groupnorm method)": [[78, "tensorrt_llm.layers.normalization.GroupNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.layernorm method)": [[78, "tensorrt_llm.layers.normalization.LayerNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.rmsnorm method)": [[78, "tensorrt_llm.layers.normalization.RmsNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.sd35adalayernormzerox method)": [[78, "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX.forward", false]], "forward() (tensorrt_llm.layers.pooling.avgpool2d method)": [[78, "tensorrt_llm.layers.pooling.AvgPool2d.forward", false]], "forward() (tensorrt_llm.models.bertforquestionanswering method)": [[79, "tensorrt_llm.models.BertForQuestionAnswering.forward", false]], "forward() (tensorrt_llm.models.bertforsequenceclassification method)": [[79, "tensorrt_llm.models.BertForSequenceClassification.forward", false]], "forward() (tensorrt_llm.models.bertmodel method)": [[79, "tensorrt_llm.models.BertModel.forward", false]], "forward() (tensorrt_llm.models.bloommodel method)": [[79, "tensorrt_llm.models.BloomModel.forward", false]], "forward() (tensorrt_llm.models.chatglmmodel method)": [[79, "tensorrt_llm.models.ChatGLMModel.forward", false]], "forward() (tensorrt_llm.models.clipvisiontransformer method)": [[79, "tensorrt_llm.models.CLIPVisionTransformer.forward", false]], "forward() (tensorrt_llm.models.decodermodel method)": [[79, "tensorrt_llm.models.DecoderModel.forward", false]], "forward() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.forward", false]], "forward() (tensorrt_llm.models.eagleforcausallm method)": [[79, "tensorrt_llm.models.EagleForCausalLM.forward", false]], "forward() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.forward", false]], "forward() (tensorrt_llm.models.falconmodel method)": [[79, "tensorrt_llm.models.FalconModel.forward", false]], "forward() (tensorrt_llm.models.gptjmodel method)": [[79, "tensorrt_llm.models.GPTJModel.forward", false]], "forward() (tensorrt_llm.models.gptmodel method)": [[79, "tensorrt_llm.models.GPTModel.forward", false]], "forward() (tensorrt_llm.models.gptneoxmodel method)": [[79, "tensorrt_llm.models.GPTNeoXModel.forward", false]], "forward() (tensorrt_llm.models.llamamodel method)": [[79, "tensorrt_llm.models.LLaMAModel.forward", false]], "forward() (tensorrt_llm.models.llavanextvisionwrapper method)": [[79, "tensorrt_llm.models.LlavaNextVisionWrapper.forward", false]], "forward() (tensorrt_llm.models.mambaforcausallm method)": [[79, "tensorrt_llm.models.MambaForCausalLM.forward", false]], "forward() (tensorrt_llm.models.mllamaforcausallm method)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM.forward", false]], "forward() (tensorrt_llm.models.mptmodel method)": [[79, "tensorrt_llm.models.MPTModel.forward", false]], "forward() (tensorrt_llm.models.optmodel method)": [[79, "tensorrt_llm.models.OPTModel.forward", false]], "forward() (tensorrt_llm.models.phi3model method)": [[79, "tensorrt_llm.models.Phi3Model.forward", false]], "forward() (tensorrt_llm.models.phimodel method)": [[79, "tensorrt_llm.models.PhiModel.forward", false]], "forward() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[79, "tensorrt_llm.models.RecurrentGemmaForCausalLM.forward", false]], "forward() (tensorrt_llm.models.redrafterforcausallm method)": [[79, "tensorrt_llm.models.ReDrafterForCausalLM.forward", false]], "forward() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.forward", false]], "forward() (tensorrt_llm.models.whisperencoder method)": [[79, "tensorrt_llm.models.WhisperEncoder.forward", false]], "forward_with_cfg() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.forward_with_cfg", false]], "forward_without_cfg() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.forward_without_cfg", false]], "fp8 (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.FP8", false]], "fp8_block_scales (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.FP8_BLOCK_SCALES", false]], "fp8_per_channel_per_token (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.FP8_PER_CHANNEL_PER_TOKEN", false]], "free_gpu_memory_fraction (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.free_gpu_memory_fraction", false]], "frequency_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.frequency_penalty", false]], "frequency_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.frequency_penalty", false]], "from_arguments() (tensorrt_llm.models.speculativedecodingmode static method)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.from_arguments", false]], "from_checkpoint() (tensorrt_llm.models.pretrainedconfig class method)": [[79, "tensorrt_llm.models.PretrainedConfig.from_checkpoint", false]], "from_checkpoint() (tensorrt_llm.models.pretrainedmodel class method)": [[79, "tensorrt_llm.models.PretrainedModel.from_checkpoint", false]], "from_config() (tensorrt_llm.models.pretrainedmodel class method)": [[79, "tensorrt_llm.models.PretrainedModel.from_config", false]], "from_dict() (tensorrt_llm.llmapi.buildconfig class method)": [[65, "tensorrt_llm.llmapi.BuildConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.calibconfig class method)": [[65, "tensorrt_llm.llmapi.CalibConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.eagledecodingconfig class method)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.lookaheaddecodingconfig class method)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.medusadecodingconfig class method)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.mtpdecodingconfig class method)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.quantconfig class method)": [[65, "tensorrt_llm.llmapi.QuantConfig.from_dict", false]], "from_dict() (tensorrt_llm.models.pretrainedconfig class method)": [[79, "tensorrt_llm.models.PretrainedConfig.from_dict", false]], "from_dir() (tensorrt_llm.runtime.modelrunner class method)": [[82, "tensorrt_llm.runtime.ModelRunner.from_dir", false]], "from_dir() (tensorrt_llm.runtime.modelrunnercpp class method)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.from_dir", false]], "from_engine() (tensorrt_llm.runtime.encdecmodelrunner class method)": [[82, "tensorrt_llm.runtime.EncDecModelRunner.from_engine", false]], "from_engine() (tensorrt_llm.runtime.modelrunner class method)": [[82, "tensorrt_llm.runtime.ModelRunner.from_engine", false]], "from_engine() (tensorrt_llm.runtime.session static method)": [[82, "tensorrt_llm.runtime.Session.from_engine", false]], "from_hugging_face() (tensorrt_llm.models.baichuanforcausallm class method)": [[79, "tensorrt_llm.models.BaichuanForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.chatglmconfig class method)": [[79, "tensorrt_llm.models.ChatGLMConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.chatglmforcausallm class method)": [[79, "tensorrt_llm.models.ChatGLMForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.cogvlmforcausallm class method)": [[79, "tensorrt_llm.models.CogVLMForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.cohereforcausallm class method)": [[79, "tensorrt_llm.models.CohereForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.deepseekforcausallm class method)": [[79, "tensorrt_llm.models.DeepseekForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.deepseekv2forcausallm class method)": [[79, "tensorrt_llm.models.DeepseekV2ForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.eagleforcausallm class method)": [[79, "tensorrt_llm.models.EagleForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.falconconfig class method)": [[79, "tensorrt_llm.models.FalconConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.falconforcausallm class method)": [[79, "tensorrt_llm.models.FalconForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gemmaconfig class method)": [[79, "tensorrt_llm.models.GemmaConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gemmaforcausallm class method)": [[79, "tensorrt_llm.models.GemmaForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptconfig class method)": [[79, "tensorrt_llm.models.GPTConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptforcausallm class method)": [[79, "tensorrt_llm.models.GPTForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptjconfig class method)": [[79, "tensorrt_llm.models.GPTJConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptjforcausallm class method)": [[79, "tensorrt_llm.models.GPTJForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llamaconfig class method)": [[79, "tensorrt_llm.models.LLaMAConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llamaforcausallm class method)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llavanextvisionconfig class method)": [[79, "tensorrt_llm.models.LlavaNextVisionConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llavanextvisionwrapper class method)": [[79, "tensorrt_llm.models.LlavaNextVisionWrapper.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.mambaforcausallm class method)": [[79, "tensorrt_llm.models.MambaForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.medusaconfig class method)": [[79, "tensorrt_llm.models.MedusaConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.medusaforcausallm class method)": [[79, "tensorrt_llm.models.MedusaForCausalLm.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.mllamaforcausallm class method)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.phi3forcausallm class method)": [[79, "tensorrt_llm.models.Phi3ForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.phiforcausallm class method)": [[79, "tensorrt_llm.models.PhiForCausalLM.from_hugging_face", false]], "from_json_file() (tensorrt_llm.llmapi.buildconfig class method)": [[65, "tensorrt_llm.llmapi.BuildConfig.from_json_file", false]], "from_json_file() (tensorrt_llm.models.pretrainedconfig class method)": [[79, "tensorrt_llm.models.PretrainedConfig.from_json_file", false]], "from_meta_ckpt() (tensorrt_llm.models.llamaconfig class method)": [[79, "tensorrt_llm.models.LLaMAConfig.from_meta_ckpt", false]], "from_meta_ckpt() (tensorrt_llm.models.llamaforcausallm class method)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.from_meta_ckpt", false]], "from_nemo() (tensorrt_llm.models.gptconfig class method)": [[79, "tensorrt_llm.models.GPTConfig.from_nemo", false]], "from_nemo() (tensorrt_llm.models.gptforcausallm class method)": [[79, "tensorrt_llm.models.GPTForCausalLM.from_nemo", false]], "from_pretrained() (tensorrt_llm.models.sd3transformer2dmodel class method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.from_pretrained", false]], "from_serialized_engine() (tensorrt_llm.runtime.session static method)": [[82, "tensorrt_llm.runtime.Session.from_serialized_engine", false]], "from_string() (tensorrt_llm.functional.positionembeddingtype static method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.from_string", false]], "from_string() (tensorrt_llm.functional.rotaryscalingtype static method)": [[77, "tensorrt_llm.functional.RotaryScalingType.from_string", false]], "fuse_qkv_projections() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.fuse_qkv_projections", false]], "fusedgatedmlp (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.FusedGatedMLP", false]], "fusedgatedmlp (tensorrt_llm.functional.mlptype attribute)": [[77, "tensorrt_llm.functional.MLPType.FusedGatedMLP", false]], "gatedmlp (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.GatedMLP", false]], "gatedmlp (tensorrt_llm.functional.mlptype attribute)": [[77, "tensorrt_llm.functional.MLPType.GatedMLP", false]], "gather() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gather", false]], "gather_context_logits (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.gather_context_logits", false]], "gather_generation_logits (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.gather_generation_logits", false]], "gather_last_token_logits() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gather_last_token_logits", false]], "gather_nd() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gather_nd", false]], "gegelu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gegelu", false]], "geglu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.geglu", false]], "gelu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gelu", false]], "gemm_allreduce() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gemm_allreduce", false]], "gemm_allreduce_plugin (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.gemm_allreduce_plugin", false]], "gemm_allreduce_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.gemm_allreduce_plugin", false]], "gemm_swiglu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gemm_swiglu", false]], "gemma2_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[79, "tensorrt_llm.models.GemmaConfig.GEMMA2_ADDED_FIELDS", false]], "gemma2_config() (tensorrt_llm.models.gemmaconfig method)": [[79, "tensorrt_llm.models.GemmaConfig.gemma2_config", false]], "gemma3_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[79, "tensorrt_llm.models.GemmaConfig.GEMMA3_ADDED_FIELDS", false]], "gemma3_config() (tensorrt_llm.models.gemmaconfig method)": [[79, "tensorrt_llm.models.GemmaConfig.gemma3_config", false]], "gemma_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[79, "tensorrt_llm.models.GemmaConfig.GEMMA_ADDED_FIELDS", false]], "gemmaconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GemmaConfig", false]], "gemmaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GemmaForCausalLM", false]], "generate() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.generate", false]], "generate() (tensorrt_llm.runtime.encdecmodelrunner method)": [[82, "tensorrt_llm.runtime.EncDecModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.modelrunner method)": [[82, "tensorrt_llm.runtime.ModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.modelrunnercpp method)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.generate", false]], "generate() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.qwenforcausallmgenerationsession method)": [[82, "tensorrt_llm.runtime.QWenForCausalLMGenerationSession.generate", false]], "generate_alibi_biases() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.generate_alibi_biases", false]], "generate_alibi_slopes() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.generate_alibi_slopes", false]], "generate_async() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.generate_async", false]], "generate_logn_scaling() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.generate_logn_scaling", false]], "generation_logits (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.generation_logits", false]], "generationsequence (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.GenerationSequence", false]], "generationsession (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.GenerationSession", false]], "get_1d_sincos_pos_embed_from_grid() (in module tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.get_1d_sincos_pos_embed_from_grid", false]], "get_2d_sincos_pos_embed() (in module tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.get_2d_sincos_pos_embed", false]], "get_2d_sincos_pos_embed_from_grid() (in module tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.get_2d_sincos_pos_embed_from_grid", false]], "get_audio_features() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.get_audio_features", false]], "get_batch_idx() (tensorrt_llm.runtime.generationsequence method)": [[82, "tensorrt_llm.runtime.GenerationSequence.get_batch_idx", false]], "get_block_offsets() (tensorrt_llm.runtime.kvcachemanager method)": [[82, "tensorrt_llm.runtime.KVCacheManager.get_block_offsets", false]], "get_comm() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.get_comm", false]], "get_config_group() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.get_config_group", false]], "get_context_phase_params() (tensorrt_llm.llmapi.disaggregatedparams method)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.get_context_phase_params", false]], "get_first_past_key_value() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[78, "tensorrt_llm.layers.attention.KeyValueCacheParams.get_first_past_key_value", false]], "get_hf_config() (tensorrt_llm.models.gemmaconfig static method)": [[79, "tensorrt_llm.models.GemmaConfig.get_hf_config", false]], "get_kv_cache_events() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.get_kv_cache_events", false]], "get_kv_cache_events_async() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.get_kv_cache_events_async", false]], "get_next_medusa_tokens() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.get_next_medusa_tokens", false]], "get_num_heads_kv() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.get_num_heads_kv", false]], "get_parent() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.get_parent", false]], "get_request_type() (tensorrt_llm.llmapi.disaggregatedparams method)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.get_request_type", false]], "get_rope_index() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.get_rope_index", false]], "get_seq_idx() (tensorrt_llm.runtime.generationsequence method)": [[82, "tensorrt_llm.runtime.GenerationSequence.get_seq_idx", false]], "get_stats() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.get_stats", false]], "get_stats_async() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.get_stats_async", false]], "get_timestep_embedding() (in module tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.get_timestep_embedding", false]], "get_users() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.get_users", false]], "get_visual_features() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.get_visual_features", false]], "get_weight() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.get_weight", false]], "gpt_attention() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gpt_attention", false]], "gpt_attention_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.gpt_attention_plugin", false]], "gptconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTConfig", false]], "gptforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTForCausalLM", false]], "gptjconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTJConfig", false]], "gptjforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTJForCausalLM", false]], "gptjmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTJModel", false]], "gptmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTModel", false]], "gptneoxforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTNeoXForCausalLM", false]], "gptneoxmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.GPTNeoXModel", false]], "gpu_weights_percent (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.gpu_weights_percent", false]], "grammar (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.grammar", false]], "greedy_sampling (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.greedy_sampling", false]], "group_norm() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.group_norm", false]], "group_size (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.group_size", false]], "groupnorm (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.GroupNorm", false]], "groupnorm (tensorrt_llm.functional.layernormtype attribute)": [[77, "tensorrt_llm.functional.LayerNormType.GroupNorm", false]], "gt() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.gt", false]], "guaranteed_no_evict (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[65, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.GUARANTEED_NO_EVICT", false]], "guided_decoding (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.guided_decoding", false]], "guideddecodingparams (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams", false]], "handle_per_step() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.handle_per_step", false]], "has_affine() (tensorrt_llm.functional.allreduceparams method)": [[77, "tensorrt_llm.functional.AllReduceParams.has_affine", false]], "has_bias() (tensorrt_llm.functional.allreduceparams method)": [[77, "tensorrt_llm.functional.AllReduceParams.has_bias", false]], "has_config_group() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.has_config_group", false]], "has_position_embedding (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.has_position_embedding", false]], "has_position_embedding (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.has_position_embedding", false]], "has_scale() (tensorrt_llm.functional.allreduceparams method)": [[77, "tensorrt_llm.functional.AllReduceParams.has_scale", false]], "has_token_type_embedding (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.has_token_type_embedding", false]], "has_token_type_embedding (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.has_token_type_embedding", false]], "has_zero_point (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.has_zero_point", false]], "head_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.head_size", false]], "head_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.head_size", false]], "hidden_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.hidden_size", false]], "host_cache_size (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.host_cache_size", false]], "identity() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.identity", false]], "ignore_eos (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.ignore_eos", false]], "include_stop_str_in_output (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.include_stop_str_in_output", false]], "index (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.index", false]], "index_select() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.index_select", false]], "infer_shapes() (tensorrt_llm.runtime.session method)": [[82, "tensorrt_llm.runtime.Session.infer_shapes", false]], "inflight (tensorrt_llm.llmapi.batchingtype attribute)": [[65, "tensorrt_llm.llmapi.BatchingType.INFLIGHT", false]], "init_audio_encoder() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.init_audio_encoder", false]], "init_image_encoder() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.init_image_encoder", false]], "init_llm() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.init_llm", false]], "init_processor() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.init_processor", false]], "init_tokenizer() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.init_tokenizer", false]], "input_timing_cache (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.input_timing_cache", false]], "int8 (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.INT8", false]], "int_clip() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.int_clip", false]], "interpolate() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.interpolate", false]], "is_alibi() (tensorrt_llm.functional.positionembeddingtype method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.is_alibi", false]], "is_deferred() (tensorrt_llm.functional.positionembeddingtype method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.is_deferred", false]], "is_dynamic() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.is_dynamic", false]], "is_gated_activation() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.is_gated_activation", false]], "is_gemma_2 (tensorrt_llm.models.gemmaconfig property)": [[79, "tensorrt_llm.models.GemmaConfig.is_gemma_2", false]], "is_gemma_3 (tensorrt_llm.models.gemmaconfig property)": [[79, "tensorrt_llm.models.GemmaConfig.is_gemma_3", false]], "is_medusa_mode (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.is_medusa_mode", false]], "is_module_excluded_from_quantization() (tensorrt_llm.llmapi.quantconfig method)": [[65, "tensorrt_llm.llmapi.QuantConfig.is_module_excluded_from_quantization", false]], "is_mrope() (tensorrt_llm.functional.positionembeddingtype method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.is_mrope", false]], "is_redrafter_mode (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.is_redrafter_mode", false]], "is_rope() (tensorrt_llm.functional.positionembeddingtype method)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.is_rope", false]], "is_trt_wrapper() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.is_trt_wrapper", false]], "is_valid() (tensorrt_llm.layers.attention.attentionparams method)": [[78, "tensorrt_llm.layers.attention.AttentionParams.is_valid", false]], "is_valid() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[78, "tensorrt_llm.layers.attention.KeyValueCacheParams.is_valid", false]], "is_valid_cross_attn() (tensorrt_llm.layers.attention.attentionparams method)": [[78, "tensorrt_llm.layers.attention.AttentionParams.is_valid_cross_attn", false]], "joint_attn_forward() (tensorrt_llm.layers.attention.diffusersattention method)": [[78, "tensorrt_llm.layers.attention.DiffusersAttention.joint_attn_forward", false]], "json (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.json", false]], "json_object (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.json_object", false]], "keyvaluecacheparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.KeyValueCacheParams", false]], "kv_cache_quant_algo (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.kv_cache_quant_algo", false]], "kv_cache_type (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.kv_cache_type", false]], "kv_cache_type (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.kv_cache_type", false]], "kv_cache_type (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.kv_cache_type", false]], "kv_dtype (tensorrt_llm.models.pretrainedconfig property)": [[79, "tensorrt_llm.models.PretrainedConfig.kv_dtype", false]], "kvcacheconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.KvCacheConfig", false]], "kvcachemanager (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.KVCacheManager", false]], "kvcacheretentionconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig", false]], "kvcacheretentionconfig.tokenrangeretentionconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig", false]], "labelembedding (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.LabelEmbedding", false]], "language_adapter_config (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.language_adapter_config", false]], "last_layer (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.last_layer", false]], "last_process_for_ub (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.LAST_PROCESS_FOR_UB", false]], "layer_norm() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.layer_norm", false]], "layer_quant_mode (tensorrt_llm.llmapi.quantconfig property)": [[65, "tensorrt_llm.llmapi.QuantConfig.layer_quant_mode", false]], "layer_types (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.layer_types", false]], "layernorm (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.LayerNorm", false]], "layernorm (tensorrt_llm.functional.layernormtype attribute)": [[77, "tensorrt_llm.functional.LayerNormType.LayerNorm", false]], "layernormpositiontype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.LayerNormPositionType", false]], "layernormtype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.LayerNormType", false]], "learned_absolute (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.learned_absolute", false]], "length (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.length", false]], "length (tensorrt_llm.llmapi.completionoutput property)": [[65, "id2", false]], "length_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.length_penalty", false]], "length_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.length_penalty", false]], "linear (class in tensorrt_llm.layers.linear)": [[78, "tensorrt_llm.layers.linear.Linear", false]], "linear (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.linear", false]], "linearactivation (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.LinearActivation", false]], "linearapproximategelu (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.LinearApproximateGELU", false]], "linearbase (class in tensorrt_llm.layers.linear)": [[78, "tensorrt_llm.layers.linear.LinearBase", false]], "lineargeglu (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.LinearGEGLU", false]], "lineargelu (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.LinearGELU", false]], "linearswiglu (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.LinearSwiGLU", false]], "llama3 (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.llama3", false]], "llamaconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.LLaMAConfig", false]], "llamaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.LLaMAForCausalLM", false]], "llamamodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.LLaMAModel", false]], "llavanextvisionconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.LlavaNextVisionConfig", false]], "llavanextvisionwrapper (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.LlavaNextVisionWrapper", false]], "llm (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.LLM", false]], "llm_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.llm_engine_dir", false]], "load() (tensorrt_llm.models.pretrainedmodel method)": [[79, "tensorrt_llm.models.PretrainedModel.load", false]], "load() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.load", false]], "load_test_audio() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.load_test_audio", false]], "load_test_data() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.load_test_data", false]], "locate_accepted_draft_tokens() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.locate_accepted_draft_tokens", false]], "location (tensorrt_llm.functional.tensor property)": [[77, "tensorrt_llm.functional.Tensor.location", false]], "log() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.log", false]], "log() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.log", false]], "log_softmax() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.log_softmax", false]], "logits_processor (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.logits_processor", false]], "logitsprocessor (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.LogitsProcessor", false]], "logitsprocessorlist (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.LogitsProcessorList", false]], "logprobs (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.logprobs", false]], "logprobs (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.logprobs", false]], "logprobs_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.logprobs_diff", false]], "logprobs_diff (tensorrt_llm.llmapi.completionoutput property)": [[65, "id3", false]], "long_rope (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.long_rope", false]], "longrope (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.longrope", false]], "lookahead_config (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.lookahead_config", false]], "lookahead_decoding (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.LOOKAHEAD_DECODING", false]], "lookaheaddecodingconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig", false]], "lora_config (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.lora_config", false]], "lora_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.lora_plugin", false]], "lora_plugin() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.lora_plugin", false]], "lora_target_modules (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.lora_target_modules", false]], "low_latency_gemm() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.low_latency_gemm", false]], "low_latency_gemm_swiglu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.low_latency_gemm_swiglu", false]], "lt() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.lt", false]], "make_causal_mask() (in module tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.make_causal_mask", false]], "mamba_conv1d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.mamba_conv1d", false]], "mamba_conv1d_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.mamba_conv1d_plugin", false]], "mambaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MambaForCausalLM", false]], "mapping (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.mapping", false]], "mapping (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.mapping", false]], "mark_output() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.mark_output", false]], "masked_scatter() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.masked_scatter", false]], "masked_select() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.masked_select", false]], "matmul() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.matmul", false]], "max() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.max", false]], "max() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.max", false]], "max_attention_window (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.max_attention_window", false]], "max_attention_window_size (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.max_attention_window_size", false]], "max_batch_size (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_batch_size", false]], "max_batch_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.max_batch_size", false]], "max_beam_width (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_beam_width", false]], "max_beam_width (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.max_beam_width", false]], "max_cache_storage_gb (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildCacheConfig.max_cache_storage_gb", false]], "max_cache_storage_gb (tensorrt_llm.llmapi.buildcacheconfig property)": [[65, "id8", false]], "max_draft_len (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_draft_len", false]], "max_draft_tokens (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.max_draft_tokens", false]], "max_encoder_input_len (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_encoder_input_len", false]], "max_input_len (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_input_len", false]], "max_medusa_tokens (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.max_medusa_tokens", false]], "max_new_tokens (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.max_new_tokens", false]], "max_ngram_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_ngram_size", false]], "max_non_leaves_per_layer (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.max_non_leaves_per_layer", false]], "max_num_tokens (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_num_tokens", false]], "max_num_tokens (tensorrt_llm.llmapi.cachetransceiverconfig attribute)": [[65, "tensorrt_llm.llmapi.CacheTransceiverConfig.max_num_tokens", false]], "max_prompt_embedding_table_size (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.max_prompt_embedding_table_size", false]], "max_records (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildCacheConfig.max_records", false]], "max_records (tensorrt_llm.llmapi.buildcacheconfig property)": [[65, "id9", false]], "max_seq_len (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.max_seq_len", false]], "max_sequence_length (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.max_sequence_length", false]], "max_sequence_length (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.max_sequence_length", false]], "max_tokens (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.max_tokens", false]], "max_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.max_tokens", false]], "max_utilization (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[65, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.MAX_UTILIZATION", false]], "max_verification_set_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_verification_set_size", false]], "max_window_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_window_size", false]], "maximum() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.maximum", false]], "mean() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.mean", false]], "mean() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.mean", false]], "medusa (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.MEDUSA", false]], "medusa_choices (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig.medusa_choices", false]], "medusa_decode_and_verify() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_decode_and_verify", false]], "medusa_paths (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_paths", false]], "medusa_position_offsets (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_position_offsets", false]], "medusa_temperature (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_temperature", false]], "medusa_topks (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_topks", false]], "medusa_tree_ids (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.medusa_tree_ids", false]], "medusaconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MedusaConfig", false]], "medusadecodingconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig", false]], "medusaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MedusaForCausalLm", false]], "meshgrid2d() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.meshgrid2d", false]], "min() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.min", false]], "min_latency (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.MIN_LATENCY", false]], "min_length (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.min_length", false]], "min_p (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.min_p", false]], "min_p (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.min_p", false]], "min_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.min_tokens", false]], "minimum() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.minimum", false]], "mish (class in tensorrt_llm.layers.activation)": [[78, "tensorrt_llm.layers.activation.Mish", false]], "mixed_precision (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.MIXED_PRECISION", false]], "mllamaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM", false]], "mlp (class in tensorrt_llm.layers.mlp)": [[78, "tensorrt_llm.layers.mlp.MLP", false]], "mlp (tensorrt_llm.functional.mlptype attribute)": [[77, "tensorrt_llm.functional.MLPType.MLP", false]], "mlptype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.MLPType", false]], "model": [[26, "cmdoption-trtllm-serve-serve-arg-MODEL", false]], "model_config (tensorrt_llm.llmapi.cachetransceiverconfig attribute)": [[65, "tensorrt_llm.llmapi.CacheTransceiverConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[65, "tensorrt_llm.llmapi.DynamicBatchConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.schedulerconfig attribute)": [[65, "tensorrt_llm.llmapi.SchedulerConfig.model_config", false]], "model_name (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.model_name", false]], "modelconfig (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.ModelConfig", false]], "modelrunner (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.ModelRunner", false]], "modelrunnercpp (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp", false]], "module": [[77, "module-tensorrt_llm", false], [77, "module-tensorrt_llm.functional", false], [78, "module-tensorrt_llm", false], [78, "module-tensorrt_llm.layers.activation", false], [78, "module-tensorrt_llm.layers.attention", false], [78, "module-tensorrt_llm.layers.cast", false], [78, "module-tensorrt_llm.layers.conv", false], [78, "module-tensorrt_llm.layers.embedding", false], [78, "module-tensorrt_llm.layers.linear", false], [78, "module-tensorrt_llm.layers.mlp", false], [78, "module-tensorrt_llm.layers.normalization", false], [78, "module-tensorrt_llm.layers.pooling", false], [79, "module-tensorrt_llm", false], [79, "module-tensorrt_llm.models", false], [80, "module-tensorrt_llm", false], [80, "module-tensorrt_llm.plugin", false], [81, "module-tensorrt_llm", false], [81, "module-tensorrt_llm.quantization", false], [82, "module-tensorrt_llm", false], [82, "module-tensorrt_llm.runtime", false]], "modulo() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.modulo", false]], "moe (tensorrt_llm.functional.sidestreamidtype attribute)": [[77, "tensorrt_llm.functional.SideStreamIDType.moe", false]], "moe_allreduce_residual_rms_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.MOE_ALLREDUCE_RESIDUAL_RMS_NORM", false]], "monitor_memory (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.monitor_memory", false]], "mpicommsession (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.MpiCommSession", false]], "mptforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MPTForCausalLM", false]], "mptmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.MPTModel", false]], "mrope (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.mrope", false]], "mrope (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.mrope", false]], "mropeparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.MropeParams", false]], "mtpdecodingconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig", false]], "mul() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.mul", false]], "multi_block_mode (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[65, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.multi_block_mode", false]], "multimodalmodelrunner (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner", false]], "multiply_and_lora() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.multiply_and_lora", false]], "multiply_collect() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.multiply_collect", false]], "multiply_collect() (tensorrt_llm.layers.linear.rowlinear method)": [[78, "tensorrt_llm.layers.linear.RowLinear.multiply_collect", false]], "n (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.n", false]], "name (tensorrt_llm.functional.tensor property)": [[77, "tensorrt_llm.functional.Tensor.name", false]], "name (tensorrt_llm.runtime.tensorinfo attribute)": [[82, "tensorrt_llm.runtime.TensorInfo.name", false]], "native_quant_flow (tensorrt_llm.models.gemmaforcausallm attribute)": [[79, "tensorrt_llm.models.GemmaForCausalLM.NATIVE_QUANT_FLOW", false]], "nccl (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.NCCL", false]], "ndim() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.ndim", false]], "network (tensorrt_llm.functional.tensor property)": [[77, "tensorrt_llm.functional.Tensor.network", false]], "next_medusa_input_ids() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.next_medusa_input_ids", false]], "no_quant (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.NO_QUANT", false]], "no_repeat_ngram_size (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.no_repeat_ngram_size", false]], "no_repeat_ngram_size (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.no_repeat_ngram_size", false]], "non_gated_version() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.non_gated_version", false]], "none (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.NONE", false]], "none (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.none", false]], "none (tensorrt_llm.models.speculativedecodingmode attribute)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode.NONE", false]], "nonzero() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.nonzero", false]], "not_op() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.not_op", false]], "num_beams (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.num_beams", false]], "num_draft_tokens (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.num_draft_tokens", false]], "num_eagle_layers (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.num_eagle_layers", false]], "num_heads (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.num_heads", false]], "num_kv_heads (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_kv_heads", false]], "num_kv_heads_per_cross_attn_layer (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_kv_heads_per_cross_attn_layer", false]], "num_kv_heads_per_layer (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_kv_heads_per_layer", false]], "num_layers (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.num_layers", false]], "num_medusa_heads (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MedusaDecodingConfig.num_medusa_heads", false]], "num_medusa_heads (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.num_medusa_heads", false]], "num_medusa_heads (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.num_medusa_heads", false]], "num_nextn_predict_layers (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.num_nextn_predict_layers", false]], "num_return_sequences (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.num_return_sequences", false]], "numel() (tensorrt_llm.runtime.tensorinfo method)": [[82, "tensorrt_llm.runtime.TensorInfo.numel", false]], "nvfp4 (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.NVFP4", false]], "nvinfer1 (c++ type)": [[1, "_CPPv48nvinfer1", false]], "onboard_blocks (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.onboard_blocks", false]], "oneshot (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.ONESHOT", false]], "op_and() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.op_and", false]], "op_or() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.op_or", false]], "op_xor() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.op_xor", false]], "opaque_state (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.opaque_state", false]], "opt_batch_size (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.opt_batch_size", false]], "opt_num_tokens (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.opt_num_tokens", false]], "optforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.OPTForCausalLM", false]], "optmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.OPTModel", false]], "outer() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.outer", false]], "output_cum_log_probs (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.output_cum_log_probs", false]], "output_log_probs (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.output_log_probs", false]], "output_sequence_lengths (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.output_sequence_lengths", false]], "output_timing_cache (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.output_timing_cache", false]], "outputs (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.outputs", false]], "pad() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.pad", false]], "pad_id (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.pad_id", false]], "pad_id (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.pad_id", false]], "padding (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.padding", false]], "paged_kv_cache (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.paged_kv_cache", false]], "paged_state (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.paged_state", false]], "paged_state (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.paged_state", false]], "permute() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.permute", false]], "permute() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.permute", false]], "phi3forcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.Phi3ForCausalLM", false]], "phi3model (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.Phi3Model", false]], "phiforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.PhiForCausalLM", false]], "phimodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.PhiModel", false]], "pixartalphatextprojection (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection", false]], "plugin_config (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.plugin_config", false]], "pluginconfig (class in tensorrt_llm.plugin)": [[80, "tensorrt_llm.plugin.PluginConfig", false]], "positionembeddingtype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.PositionEmbeddingType", false]], "post_layernorm (tensorrt_llm.functional.layernormpositiontype attribute)": [[77, "tensorrt_llm.functional.LayerNormPositionType.post_layernorm", false]], "posterior_threshold (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.posterior_threshold", false]], "postprocess() (tensorrt_llm.layers.attention.attention method)": [[78, "tensorrt_llm.layers.attention.Attention.postprocess", false]], "postprocess() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[78, "tensorrt_llm.layers.attention.DeepseekV2Attention.postprocess", false]], "postprocess() (tensorrt_llm.layers.embedding.embedding method)": [[78, "tensorrt_llm.layers.embedding.Embedding.postprocess", false]], "postprocess() (tensorrt_llm.layers.linear.linear method)": [[78, "tensorrt_llm.layers.linear.Linear.postprocess", false]], "pow() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.pow", false]], "pp_communicate_final_output_ids() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.pp_communicate_final_output_ids", false]], "pp_communicate_new_tokens() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.pp_communicate_new_tokens", false]], "pre_layernorm (tensorrt_llm.functional.layernormpositiontype attribute)": [[77, "tensorrt_llm.functional.LayerNormPositionType.pre_layernorm", false]], "pre_quant_scale (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.pre_quant_scale", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.decodermodel method)": [[79, "tensorrt_llm.models.DecoderModel.precompute_relative_attention_bias", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.precompute_relative_attention_bias", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.whisperencoder method)": [[79, "tensorrt_llm.models.WhisperEncoder.precompute_relative_attention_bias", false]], "prepare_inputs() (tensorrt_llm.models.chatglmforcausallm method)": [[79, "tensorrt_llm.models.ChatGLMForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.decodermodel method)": [[79, "tensorrt_llm.models.DecoderModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.eagleforcausallm method)": [[79, "tensorrt_llm.models.EagleForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.llavanextvisionwrapper method)": [[79, "tensorrt_llm.models.LlavaNextVisionWrapper.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.mambaforcausallm method)": [[79, "tensorrt_llm.models.MambaForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.mllamaforcausallm method)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.pretrainedmodel method)": [[79, "tensorrt_llm.models.PretrainedModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[79, "tensorrt_llm.models.RecurrentGemmaForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.redrafterforcausallm method)": [[79, "tensorrt_llm.models.ReDrafterForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.whisperencoder method)": [[79, "tensorrt_llm.models.WhisperEncoder.prepare_inputs", false]], "prepare_position_ids_for_cogvlm() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.prepare_position_ids_for_cogvlm", false]], "prepare_recurrent_inputs() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[79, "tensorrt_llm.models.RecurrentGemmaForCausalLM.prepare_recurrent_inputs", false]], "preprocess() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.preprocess", false]], "presence_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.presence_penalty", false]], "presence_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.presence_penalty", false]], "pretrainedconfig (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.PretrainedConfig", false]], "pretrainedmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.PretrainedModel", false]], "priority (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.priority", false]], "process_input() (tensorrt_llm.runtime.encdecmodelrunner method)": [[82, "tensorrt_llm.runtime.EncDecModelRunner.process_input", false]], "process_logits_including_draft() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.process_logits_including_draft", false]], "prod() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.prod", false]], "profiler (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.profiler", false]], "profiling_verbosity (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.profiling_verbosity", false]], "prompt (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.prompt", false]], "prompt (tensorrt_llm.llmapi.requestoutput property)": [[65, "id6", false]], "prompt_logprobs (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.prompt_logprobs", false]], "prompt_logprobs (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.prompt_logprobs", false]], "prompt_token_ids (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.prompt_token_ids", false]], "prompttuningembedding (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.PromptTuningEmbedding", false]], "ptuning_setup() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup", false]], "ptuning_setup_fuyu() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_fuyu", false]], "ptuning_setup_llava_next() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_llava_next", false]], "ptuning_setup_phi3() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_phi3", false]], "ptuning_setup_pixtral() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_pixtral", false]], "python_e2e (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.python_e2e", false]], "pytorch_eagle_weights_path (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.pytorch_eagle_weights_path", false]], "quant_algo (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.quant_algo", false]], "quant_algo (tensorrt_llm.models.pretrainedconfig property)": [[79, "tensorrt_llm.models.PretrainedConfig.quant_algo", false]], "quant_mode (tensorrt_llm.llmapi.quantconfig property)": [[65, "tensorrt_llm.llmapi.QuantConfig.quant_mode", false]], "quant_mode (tensorrt_llm.models.pretrainedconfig property)": [[79, "tensorrt_llm.models.PretrainedConfig.quant_mode", false]], "quant_mode (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.quant_mode", false]], "quant_mode (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.quant_mode", false]], "quantalgo (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.QuantAlgo", false]], "quantalgo (class in tensorrt_llm.quantization)": [[81, "tensorrt_llm.quantization.QuantAlgo", false]], "quantconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.QuantConfig", false]], "quantize() (tensorrt_llm.models.baichuanforcausallm class method)": [[79, "tensorrt_llm.models.BaichuanForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.chatglmforcausallm class method)": [[79, "tensorrt_llm.models.ChatGLMForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.cogvlmforcausallm class method)": [[79, "tensorrt_llm.models.CogVLMForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.gemmaforcausallm class method)": [[79, "tensorrt_llm.models.GemmaForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.gptforcausallm class method)": [[79, "tensorrt_llm.models.GPTForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.llamaforcausallm class method)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.pretrainedmodel class method)": [[79, "tensorrt_llm.models.PretrainedModel.quantize", false]], "quantize_and_export() (in module tensorrt_llm.quantization)": [[81, "tensorrt_llm.quantization.quantize_and_export", false]], "quantmode (class in tensorrt_llm.quantization)": [[81, "tensorrt_llm.quantization.QuantMode", false]], "quick_gelu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.quick_gelu", false]], "qwenforcausallmgenerationsession (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.QWenForCausalLMGenerationSession", false]], "rand() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.rand", false]], "random_seed (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.random_seed", false]], "random_seed (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.random_seed", false]], "rank() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.rank", false]], "rearrange() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.rearrange", false]], "recurrentgemmaforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.RecurrentGemmaForCausalLM", false]], "recv() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.recv", false]], "redrafter_draft_len_per_beam (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.redrafter_draft_len_per_beam", false]], "redrafter_num_beams (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.redrafter_num_beams", false]], "redrafterforcausallm (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.ReDrafterForCausalLM", false]], "reduce() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.reduce", false]], "reduce_scatter() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.reduce_scatter", false]], "regex (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.regex", false]], "relative (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.relative", false]], "relaxed_delta (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.relaxed_delta", false]], "relaxed_topk (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.relaxed_topk", false]], "release() (tensorrt_llm.models.pretrainedmodel method)": [[79, "tensorrt_llm.models.PretrainedModel.release", false]], "relu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.relu", false]], "remove_input_padding (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.remove_input_padding", false]], "reorder_kv_cache_for_beam_search() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.reorder_kv_cache_for_beam_search", false]], "repeat() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.repeat", false]], "repeat() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.repeat", false]], "repeat_interleave() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.repeat_interleave", false]], "repetition_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.repetition_penalty", false]], "repetition_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.repetition_penalty", false]], "replace_all_uses_with() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.replace_all_uses_with", false]], "request_id (tensorrt_llm.llmapi.requestoutput attribute)": [[65, "tensorrt_llm.llmapi.RequestOutput.request_id", false]], "request_type (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[65, "tensorrt_llm.llmapi.DisaggregatedParams.request_type", false]], "requesterror (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.RequestError", false]], "requestoutput (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.RequestOutput", false]], "residual_rms_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM", false]], "residual_rms_norm_out_quant_fp8 (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_OUT_QUANT_FP8", false]], "residual_rms_norm_out_quant_nvfp4 (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_OUT_QUANT_NVFP4", false]], "residual_rms_norm_quant_fp8 (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_QUANT_FP8", false]], "residual_rms_norm_quant_nvfp4 (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_QUANT_NVFP4", false]], "residual_rms_prepost_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[77, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_PREPOST_NORM", false]], "return_context_logits (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.return_context_logits", false]], "return_dict (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.return_dict", false]], "return_encoder_output (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.return_encoder_output", false]], "return_generation_logits (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.return_generation_logits", false]], "return_perf_metrics (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.return_perf_metrics", false]], "rg_lru() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.rg_lru", false]], "rms_norm() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.rms_norm", false]], "rmsnorm (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.RmsNorm", false]], "rmsnorm (tensorrt_llm.functional.layernormtype attribute)": [[77, "tensorrt_llm.functional.LayerNormType.RmsNorm", false]], "rnn_conv_dim_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.rnn_conv_dim_size", false]], "rnn_conv_dim_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.rnn_conv_dim_size", false]], "rnn_head_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.rnn_head_size", false]], "rnn_head_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.rnn_head_size", false]], "rnn_hidden_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.rnn_hidden_size", false]], "rnn_hidden_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.rnn_hidden_size", false]], "robertaforquestionanswering (in module tensorrt_llm.models)": [[79, "tensorrt_llm.models.RobertaForQuestionAnswering", false]], "robertaforsequenceclassification (in module tensorrt_llm.models)": [[79, "tensorrt_llm.models.RobertaForSequenceClassification", false]], "robertamodel (in module tensorrt_llm.models)": [[79, "tensorrt_llm.models.RobertaModel", false]], "rope_gpt_neox (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.rope_gpt_neox", false]], "rope_gptj (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.rope_gptj", false]], "ropeembeddingutils (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils", false]], "rotaryscalingtype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.RotaryScalingType", false]], "rotate_every_two() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.rotate_every_two", false]], "rotate_half() (tensorrt_llm.functional.ropeembeddingutils static method)": [[77, "tensorrt_llm.functional.RopeEmbeddingUtils.rotate_half", false]], "round() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.round", false]], "rowlinear (class in tensorrt_llm.layers.linear)": [[78, "tensorrt_llm.layers.linear.RowLinear", false]], "run() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.run", false]], "run() (tensorrt_llm.runtime.session method)": [[82, "tensorrt_llm.runtime.Session.run", false]], "runtime (tensorrt_llm.runtime.generationsession attribute)": [[82, "tensorrt_llm.runtime.GenerationSession.runtime", false]], "runtime (tensorrt_llm.runtime.session property)": [[82, "tensorrt_llm.runtime.Session.runtime", false]], "samplingconfig (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.SamplingConfig", false]], "samplingparams (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.SamplingParams", false]], "save() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.save", false]], "save_checkpoint() (tensorrt_llm.models.llavanextvisionwrapper method)": [[79, "tensorrt_llm.models.LlavaNextVisionWrapper.save_checkpoint", false]], "save_checkpoint() (tensorrt_llm.models.pretrainedmodel method)": [[79, "tensorrt_llm.models.PretrainedModel.save_checkpoint", false]], "scatter() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.scatter", false]], "scatter_nd() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.scatter_nd", false]], "schedulerconfig (class in tensorrt_llm.llmapi)": [[65, "tensorrt_llm.llmapi.SchedulerConfig", false]], "sd35adalayernormzerox (class in tensorrt_llm.layers.normalization)": [[78, "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX", false]], "sd3patchembed (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.SD3PatchEmbed", false]], "sd3transformer2dmodel (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.SD3Transformer2DModel", false]], "secondary_offload_min_priority (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.secondary_offload_min_priority", false]], "seed (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.seed", false]], "select() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.select", false]], "select() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.select", false]], "selective_scan() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.selective_scan", false]], "send() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.send", false]], "serialize_engine() (tensorrt_llm.runtime.modelrunner method)": [[82, "tensorrt_llm.runtime.ModelRunner.serialize_engine", false]], "session (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.Session", false]], "set_attn_processor() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.set_attn_processor", false]], "set_from_optional (c macro)": [[1, "c.SET_FROM_OPTIONAL", false]], "set_if_not_exist() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.set_if_not_exist", false]], "set_rank() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.set_rank", false]], "set_rel_attn_table() (tensorrt_llm.layers.attention.attention method)": [[78, "tensorrt_llm.layers.attention.Attention.set_rel_attn_table", false]], "set_shapes() (tensorrt_llm.runtime.session method)": [[82, "tensorrt_llm.runtime.Session.set_shapes", false]], "setup() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.setup", false]], "setup_fake_prompts() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts", false]], "setup_fake_prompts_qwen2vl() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts_qwen2vl", false]], "setup_fake_prompts_vila() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts_vila", false]], "setup_inputs() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.setup_inputs", false]], "shape (tensorrt_llm.functional.tensor property)": [[77, "tensorrt_llm.functional.Tensor.shape", false]], "shape (tensorrt_llm.runtime.tensorinfo attribute)": [[82, "tensorrt_llm.runtime.TensorInfo.shape", false]], "shape() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.shape", false]], "shutdown() (tensorrt_llm.llmapi.llm method)": [[65, "tensorrt_llm.llmapi.LLM.shutdown", false]], "shutdown() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.shutdown", false]], "sidestreamidtype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.SideStreamIDType", false]], "sigmoid() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.sigmoid", false]], "silu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.silu", false]], "sin() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.sin", false]], "sink_token_length (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[65, "tensorrt_llm.llmapi.KvCacheConfig.sink_token_length", false]], "sink_token_length (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.sink_token_length", false]], "size (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.size", false]], "size() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.size", false]], "skip_cross_attn_blocks (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.skip_cross_attn_blocks", false]], "skip_cross_kv (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.skip_cross_kv", false]], "skip_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.skip_special_tokens", false]], "slice() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.slice", false]], "sliceinputtype (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.SliceInputType", false]], "sliding_window_causal (tensorrt_llm.functional.attentionmasktype attribute)": [[77, "tensorrt_llm.functional.AttentionMaskType.sliding_window_causal", false]], "smoothquant_val (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.smoothquant_val", false]], "softmax() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.softmax", false]], "softplus() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.softplus", false]], "spaces_between_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.spaces_between_special_tokens", false]], "specdecodingparams (class in tensorrt_llm.layers.attention)": [[78, "tensorrt_llm.layers.attention.SpecDecodingParams", false]], "speculative_decoding_mode (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.speculative_decoding_mode", false]], "speculativedecodingmode (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.SpeculativeDecodingMode", false]], "split() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.split", false]], "split() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.split", false]], "split_prompt_by_images() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.split_prompt_by_images", false]], "sqrt() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.sqrt", false]], "sqrt() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.sqrt", false]], "squared_relu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.squared_relu", false]], "squeeze() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.squeeze", false]], "squeeze() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.squeeze", false]], "squeeze() (tensorrt_llm.runtime.tensorinfo method)": [[82, "tensorrt_llm.runtime.TensorInfo.squeeze", false]], "stack() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.stack", false]], "start (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.start", false]], "state_dtype (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.state_dtype", false]], "state_dtype (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.state_dtype", false]], "state_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.state_size", false]], "state_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.state_size", false]], "static (tensorrt_llm.llmapi.batchingtype attribute)": [[65, "tensorrt_llm.llmapi.BatchingType.STATIC", false]], "static_batch (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[65, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.STATIC_BATCH", false]], "step() (tensorrt_llm.runtime.kvcachemanager method)": [[82, "tensorrt_llm.runtime.KVCacheManager.step", false]], "stop (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.stop", false]], "stop_reason (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.stop_reason", false]], "stop_token_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.stop_token_ids", false]], "stop_words_list (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.stop_words_list", false]], "stoppingcriteria (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.StoppingCriteria", false]], "stoppingcriterialist (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.StoppingCriteriaList", false]], "stride (tensorrt_llm.functional.sliceinputtype attribute)": [[77, "tensorrt_llm.functional.SliceInputType.stride", false]], "strongly_typed (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.strongly_typed", false]], "structural_tag (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[65, "tensorrt_llm.llmapi.GuidedDecodingParams.structural_tag", false]], "sub() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.sub", false]], "submit() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.submit", false]], "submit_sync() (tensorrt_llm.llmapi.mpicommsession method)": [[65, "tensorrt_llm.llmapi.MpiCommSession.submit_sync", false]], "sum() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.sum", false]], "swiglu() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.swiglu", false]], "tanh() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.tanh", false]], "temperature (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.temperature", false]], "temperature (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.temperature", false]], "tensor (class in tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.Tensor", false]], "tensorinfo (class in tensorrt_llm.runtime)": [[82, "tensorrt_llm.runtime.TensorInfo", false]], "tensorrt_llm": [[77, "module-tensorrt_llm", false], [78, "module-tensorrt_llm", false], [79, "module-tensorrt_llm", false], [80, "module-tensorrt_llm", false], [81, "module-tensorrt_llm", false], [82, "module-tensorrt_llm", false]], "tensorrt_llm (c++ type)": [[0, "_CPPv412tensorrt_llm", false], [1, "_CPPv412tensorrt_llm", false]], "tensorrt_llm.functional": [[77, "module-tensorrt_llm.functional", false]], "tensorrt_llm.layers.activation": [[78, "module-tensorrt_llm.layers.activation", false]], "tensorrt_llm.layers.attention": [[78, "module-tensorrt_llm.layers.attention", false]], "tensorrt_llm.layers.cast": [[78, "module-tensorrt_llm.layers.cast", false]], "tensorrt_llm.layers.conv": [[78, "module-tensorrt_llm.layers.conv", false]], "tensorrt_llm.layers.embedding": [[78, "module-tensorrt_llm.layers.embedding", false]], "tensorrt_llm.layers.linear": [[78, "module-tensorrt_llm.layers.linear", false]], "tensorrt_llm.layers.mlp": [[78, "module-tensorrt_llm.layers.mlp", false]], "tensorrt_llm.layers.normalization": [[78, "module-tensorrt_llm.layers.normalization", false]], "tensorrt_llm.layers.pooling": [[78, "module-tensorrt_llm.layers.pooling", false]], "tensorrt_llm.models": [[79, "module-tensorrt_llm.models", false]], "tensorrt_llm.plugin": [[80, "module-tensorrt_llm.plugin", false]], "tensorrt_llm.quantization": [[81, "module-tensorrt_llm.quantization", false]], "tensorrt_llm.runtime": [[82, "module-tensorrt_llm.runtime", false]], "tensorrt_llm::batch_manager (c++ type)": [[0, "_CPPv4N12tensorrt_llm13batch_managerE", false], [1, "_CPPv4N12tensorrt_llm13batch_managerE", false]], "tensorrt_llm::batch_manager::kv_cache_manager (c++ type)": [[0, "_CPPv4N12tensorrt_llm13batch_manager16kv_cache_managerE", false]], "tensorrt_llm::executor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executorE", false]], "tensorrt_llm::executor::additionalmodeloutput (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutputE", false]], "tensorrt_llm::executor::additionalmodeloutput::additionalmodeloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", false]], "tensorrt_llm::executor::additionalmodeloutput::gathercontext (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput13gatherContextE", false]], "tensorrt_llm::executor::additionalmodeloutput::name (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput4nameE", false]], "tensorrt_llm::executor::additionalmodeloutput::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", false]], "tensorrt_llm::executor::additionaloutput (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputE", false]], "tensorrt_llm::executor::additionaloutput::additionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", false]], "tensorrt_llm::executor::additionaloutput::name (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput4nameE", false]], "tensorrt_llm::executor::additionaloutput::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", false]], "tensorrt_llm::executor::additionaloutput::output (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput6outputE", false]], "tensorrt_llm::executor::additionaloutput::~additionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputD0Ev", false]], "tensorrt_llm::executor::batchingtype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingTypeE", false]], "tensorrt_llm::executor::batchingtype::kinflight (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingType9kINFLIGHTE", false]], "tensorrt_llm::executor::batchingtype::kstatic (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingType7kSTATICE", false]], "tensorrt_llm::executor::beamtokens (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10BeamTokensE", false]], "tensorrt_llm::executor::bufferview (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10BufferViewE", false]], "tensorrt_llm::executor::cachetransceiverconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfigE", false]], "tensorrt_llm::executor::cachetransceiverconfig::cachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", false]], "tensorrt_llm::executor::cachetransceiverconfig::getmaxnumtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfig15getMaxNumTokensEv", false]], "tensorrt_llm::executor::cachetransceiverconfig::mmaxnumtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig13mMaxNumTokensE", false]], "tensorrt_llm::executor::cachetransceiverconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", false]], "tensorrt_llm::executor::cachetransceiverconfig::setmaxnumtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", false]], "tensorrt_llm::executor::capacityschedulerpolicy (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicyE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kguaranteed_no_evict (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy20kGUARANTEED_NO_EVICTE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kmax_utilization (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy16kMAX_UTILIZATIONE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kstatic_batch (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy13kSTATIC_BATCHE", false]], "tensorrt_llm::executor::communicationmode (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationModeE", false]], "tensorrt_llm::executor::communicationmode::kleader (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationMode7kLEADERE", false]], "tensorrt_llm::executor::communicationmode::korchestrator (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationMode13kORCHESTRATORE", false]], "tensorrt_llm::executor::communicationtype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationTypeE", false]], "tensorrt_llm::executor::communicationtype::kmpi (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationType4kMPIE", false]], "tensorrt_llm::executor::contextchunkingpolicy (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicyE", false]], "tensorrt_llm::executor::contextchunkingpolicy::kequal_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy15kEQUAL_PROGRESSE", false]], "tensorrt_llm::executor::contextchunkingpolicy::kfirst_come_first_served (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy24kFIRST_COME_FIRST_SERVEDE", false]], "tensorrt_llm::executor::contextphaseparams (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsE", false]], "tensorrt_llm::executor::contextphaseparams::contextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERR18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::deleter (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", false]], "tensorrt_llm::executor::contextphaseparams::getdrafttokens (c++ function)": [[0, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams14getDraftTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::getfirstgentokens (c++ function)": [[0, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams17getFirstGenTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::getreqid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getReqIdEv", false]], "tensorrt_llm::executor::contextphaseparams::getserializedstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams18getSerializedStateEv", false]], "tensorrt_llm::executor::contextphaseparams::getstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8getStateEv", false], [0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getStateEv", false]], "tensorrt_llm::executor::contextphaseparams::mdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12mDraftTokensE", false]], "tensorrt_llm::executor::contextphaseparams::mfirstgentokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams15mFirstGenTokensE", false]], "tensorrt_llm::executor::contextphaseparams::mreqid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mReqIdE", false]], "tensorrt_llm::executor::contextphaseparams::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mStateE", false]], "tensorrt_llm::executor::contextphaseparams::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERR18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParamseqERK18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::popfirstgentokens (c++ function)": [[0, "_CPPv4NO12tensorrt_llm8executor18ContextPhaseParams17popFirstGenTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::releasestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12releaseStateEv", false]], "tensorrt_llm::executor::contextphaseparams::requestidtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams13RequestIdTypeE", false]], "tensorrt_llm::executor::contextphaseparams::stateptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8StatePtrE", false]], "tensorrt_llm::executor::contextphaseparams::~contextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsD0Ev", false]], "tensorrt_llm::executor::datatransceiverstate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverStateE", false]], "tensorrt_llm::executor::datatransceiverstate::datatransceiverstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", false], [0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::getcachestate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState13getCacheStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::getcommstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState12getCommStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::mcachestate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState11mCacheStateE", false]], "tensorrt_llm::executor::datatransceiverstate::mcommstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState10mCommStateE", false]], "tensorrt_llm::executor::datatransceiverstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", false]], "tensorrt_llm::executor::datatransceiverstate::setcachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::datatransceiverstate::setcommstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", false]], "tensorrt_llm::executor::datatransceiverstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState8toStringEv", false]], "tensorrt_llm::executor::datatype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8DataTypeE", false]], "tensorrt_llm::executor::datatype::kbf16 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kBF16E", false]], "tensorrt_llm::executor::datatype::kbool (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kBOOLE", false]], "tensorrt_llm::executor::datatype::kfp16 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kFP16E", false]], "tensorrt_llm::executor::datatype::kfp32 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kFP32E", false]], "tensorrt_llm::executor::datatype::kfp8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType4kFP8E", false]], "tensorrt_llm::executor::datatype::kint32 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kINT32E", false]], "tensorrt_llm::executor::datatype::kint64 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kINT64E", false]], "tensorrt_llm::executor::datatype::kint8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kINT8E", false]], "tensorrt_llm::executor::datatype::kuint8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kUINT8E", false]], "tensorrt_llm::executor::datatype::kunknown (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType8kUNKNOWNE", false]], "tensorrt_llm::executor::debugconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfigE", false]], "tensorrt_llm::executor::debugconfig::debugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", false]], "tensorrt_llm::executor::debugconfig::getdebuginputtensors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig20getDebugInputTensorsEv", false]], "tensorrt_llm::executor::debugconfig::getdebugoutputtensors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig21getDebugOutputTensorsEv", false]], "tensorrt_llm::executor::debugconfig::getdebugtensornames (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig19getDebugTensorNamesEv", false]], "tensorrt_llm::executor::debugconfig::getdebugtensorsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig28getDebugTensorsMaxIterationsEv", false]], "tensorrt_llm::executor::debugconfig::mdebuginputtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig18mDebugInputTensorsE", false]], "tensorrt_llm::executor::debugconfig::mdebugoutputtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig19mDebugOutputTensorsE", false]], "tensorrt_llm::executor::debugconfig::mdebugtensornames (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig17mDebugTensorNamesE", false]], "tensorrt_llm::executor::debugconfig::mdebugtensorsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig26mDebugTensorsMaxIterationsE", false]], "tensorrt_llm::executor::debugconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", false]], "tensorrt_llm::executor::debugconfig::setdebuginputtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", false]], "tensorrt_llm::executor::debugconfig::setdebugoutputtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", false]], "tensorrt_llm::executor::debugconfig::setdebugtensornames (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", false]], "tensorrt_llm::executor::debugconfig::setdebugtensorsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::debugconfig::stringvec (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig9StringVecE", false]], "tensorrt_llm::executor::debugtensorsperiteration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIterationE", false]], "tensorrt_llm::executor::debugtensorsperiteration::debugtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration12debugTensorsE", false]], "tensorrt_llm::executor::debugtensorsperiteration::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration4iterE", false]], "tensorrt_llm::executor::decodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfigE", false]], "tensorrt_llm::executor::decodingconfig::decodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", false]], "tensorrt_llm::executor::decodingconfig::enableseamlesslookaheaddecoding (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31enableSeamlessLookaheadDecodingEv", false]], "tensorrt_llm::executor::decodingconfig::getdecodingmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig15getDecodingModeEv", false]], "tensorrt_llm::executor::decodingconfig::geteagleconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig14getEagleConfigEv", false]], "tensorrt_llm::executor::decodingconfig::getlookaheaddecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig26getLookaheadDecodingConfigEv", false]], "tensorrt_llm::executor::decodingconfig::getlookaheaddecodingmaxnumrequest (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig33getLookaheadDecodingMaxNumRequestEv", false]], "tensorrt_llm::executor::decodingconfig::getmedusachoices (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig16getMedusaChoicesEv", false]], "tensorrt_llm::executor::decodingconfig::mdecodingmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig13mDecodingModeE", false]], "tensorrt_llm::executor::decodingconfig::meagleconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig12mEagleConfigE", false]], "tensorrt_llm::executor::decodingconfig::mlookaheaddecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig24mLookaheadDecodingConfigE", false]], "tensorrt_llm::executor::decodingconfig::mlookaheaddecodingmaxnumrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31mLookaheadDecodingMaxNumRequestE", false]], "tensorrt_llm::executor::decodingconfig::mmedusachoices (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14mMedusaChoicesE", false]], "tensorrt_llm::executor::decodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", false]], "tensorrt_llm::executor::decodingconfig::setdecodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig15setDecodingModeERK12DecodingMode", false]], "tensorrt_llm::executor::decodingconfig::seteagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14setEagleConfigERK11EagleConfig", false]], "tensorrt_llm::executor::decodingconfig::setlookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::decodingconfig::setmedusachoices (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig16setMedusaChoicesERK13MedusaChoices", false]], "tensorrt_llm::executor::decodingmode (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingModeE", false]], "tensorrt_llm::executor::decodingmode::allbitset (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::anybitset (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::auto (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4AutoEv", false]], "tensorrt_llm::executor::decodingmode::beamsearch (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode10BeamSearchEv", false]], "tensorrt_llm::executor::decodingmode::decodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::eagle (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5EagleEv", false]], "tensorrt_llm::executor::decodingmode::explicitdrafttokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExplicitDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::externaldrafttokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExternalDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::getname (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7getNameEv", false]], "tensorrt_llm::executor::decodingmode::getstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8getStateEv", false]], "tensorrt_llm::executor::decodingmode::isauto (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isAutoEv", false]], "tensorrt_llm::executor::decodingmode::isbeamsearch (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isBeamSearchEv", false]], "tensorrt_llm::executor::decodingmode::iseagle (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7isEagleEv", false]], "tensorrt_llm::executor::decodingmode::isexplicitdrafttokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExplicitDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::isexternaldrafttokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExternalDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::islookahead (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode11isLookaheadEv", false]], "tensorrt_llm::executor::decodingmode::ismedusa (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8isMedusaEv", false]], "tensorrt_llm::executor::decodingmode::istopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopKEv", false]], "tensorrt_llm::executor::decodingmode::istopkandtopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isTopKandTopPEv", false]], "tensorrt_llm::executor::decodingmode::istopkortopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isTopKorTopPEv", false]], "tensorrt_llm::executor::decodingmode::istopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopPEv", false]], "tensorrt_llm::executor::decodingmode::isusebantokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseBanTokensEv", false]], "tensorrt_llm::executor::decodingmode::isusebanwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isUseBanWordsEv", false]], "tensorrt_llm::executor::decodingmode::isuseexpliciteosstop (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUseExplicitEosStopEv", false]], "tensorrt_llm::executor::decodingmode::isusefrequencypenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isUseFrequencyPenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusemaxlengthstop (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode18isUseMaxLengthStopEv", false]], "tensorrt_llm::executor::decodingmode::isuseminlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseMinLengthEv", false]], "tensorrt_llm::executor::decodingmode::isuseminp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9isUseMinPEv", false]], "tensorrt_llm::executor::decodingmode::isusenorepeatngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseNoRepeatNgramSizeEv", false]], "tensorrt_llm::executor::decodingmode::isuseoccurrencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseOccurrencePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isUsePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusepresencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUsePresencePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isuserepetitionpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseRepetitionPenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusestopcriteria (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode17isUseStopCriteriaEv", false]], "tensorrt_llm::executor::decodingmode::isusestopwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseStopWordsEv", false]], "tensorrt_llm::executor::decodingmode::isusetemperature (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode16isUseTemperatureEv", false]], "tensorrt_llm::executor::decodingmode::isusevariablebeamwidthsearch (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode28isUseVariableBeamWidthSearchEv", false]], "tensorrt_llm::executor::decodingmode::kauto (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kAutoE", false]], "tensorrt_llm::executor::decodingmode::kbeamsearch (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode11kBeamSearchE", false]], "tensorrt_llm::executor::decodingmode::keagle (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6kEagleE", false]], "tensorrt_llm::executor::decodingmode::kexplicitdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExplicitDraftTokensE", false]], "tensorrt_llm::executor::decodingmode::kexternaldrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExternalDraftTokensE", false]], "tensorrt_llm::executor::decodingmode::klookahead (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode10kLookaheadE", false]], "tensorrt_llm::executor::decodingmode::kmedusa (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode7kMedusaE", false]], "tensorrt_llm::executor::decodingmode::knumflags (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kNumFlagsE", false]], "tensorrt_llm::executor::decodingmode::ktopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopKE", false]], "tensorrt_llm::executor::decodingmode::ktopktopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kTopKTopPE", false]], "tensorrt_llm::executor::decodingmode::ktopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopPE", false]], "tensorrt_llm::executor::decodingmode::kusebantokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseBanTokensE", false]], "tensorrt_llm::executor::decodingmode::kusebanwords (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12kUseBanWordsE", false]], "tensorrt_llm::executor::decodingmode::kuseexpliciteosstop (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19kUseExplicitEosStopE", false]], "tensorrt_llm::executor::decodingmode::kusefrequencypenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode22kUseFrequencyPenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusemaxlengthstop (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode17kUseMaxLengthStopE", false]], "tensorrt_llm::executor::decodingmode::kuseminlength (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseMinLengthE", false]], "tensorrt_llm::executor::decodingmode::kuseminp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8kUseMinPE", false]], "tensorrt_llm::executor::decodingmode::kusenorepeatngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUseNoRepeatNgramSizeE", false]], "tensorrt_llm::executor::decodingmode::kuseoccurrencepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseOccurrencePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUsePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusepresencepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUsePresencePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kuserepetitionpenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseRepetitionPenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusestandardstopcriteria (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode24kUseStandardStopCriteriaE", false]], "tensorrt_llm::executor::decodingmode::kusestopwords (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseStopWordsE", false]], "tensorrt_llm::executor::decodingmode::kusetemperature (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode15kUseTemperatureE", false]], "tensorrt_llm::executor::decodingmode::kusevariablebeamwidthsearch (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode27kUseVariableBeamWidthSearchE", false]], "tensorrt_llm::executor::decodingmode::lookahead (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9LookaheadEv", false]], "tensorrt_llm::executor::decodingmode::medusa (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6MedusaEv", false]], "tensorrt_llm::executor::decodingmode::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6mStateE", false]], "tensorrt_llm::executor::decodingmode::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", false]], "tensorrt_llm::executor::decodingmode::setbitto (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", false]], "tensorrt_llm::executor::decodingmode::topk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopKEv", false]], "tensorrt_llm::executor::decodingmode::topktopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8TopKTopPEv", false]], "tensorrt_llm::executor::decodingmode::topp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopPEv", false]], "tensorrt_llm::executor::decodingmode::underlyingtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode14UnderlyingTypeE", false]], "tensorrt_llm::executor::decodingmode::usebantokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", false]], "tensorrt_llm::executor::decodingmode::usebanwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", false]], "tensorrt_llm::executor::decodingmode::useexpliciteosstop (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", false]], "tensorrt_llm::executor::decodingmode::usefrequencypenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", false]], "tensorrt_llm::executor::decodingmode::usemaxlengthstop (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", false]], "tensorrt_llm::executor::decodingmode::useminlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", false]], "tensorrt_llm::executor::decodingmode::useminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", false]], "tensorrt_llm::executor::decodingmode::usenorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", false]], "tensorrt_llm::executor::decodingmode::useoccurrencepenalties (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", false]], "tensorrt_llm::executor::decodingmode::usepresencepenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", false]], "tensorrt_llm::executor::decodingmode::userepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", false]], "tensorrt_llm::executor::decodingmode::usestopwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", false]], "tensorrt_llm::executor::decodingmode::usetemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", false]], "tensorrt_llm::executor::decodingmode::usevariablebeamwidthsearch (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", false]], "tensorrt_llm::executor::detail (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6detailE", false]], "tensorrt_llm::executor::detail::dimtype64 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9DimType64E", false]], "tensorrt_llm::executor::detail::ofitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", false]], "tensorrt_llm::executor::detail::toitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", false]], "tensorrt_llm::executor::disagg_executor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executorE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::awaitcontextresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::awaitgenerationresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::canenqueue (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator10canEnqueueEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::disaggexecutororchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::enqueuecontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::enqueuegeneration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::getcontextexecutors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator19getContextExecutorsEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::getgenexecutors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator15getGenExecutorsEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator5mImplE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::~disaggexecutororchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorD0Ev", false]], "tensorrt_llm::executor::disagg_executor::responsewithid (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::gid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId3gidE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::response (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId8responseE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::responsewithid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::~responsewithid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdD0Ev", false]], "tensorrt_llm::executor::disservingrequeststats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStatsE", false]], "tensorrt_llm::executor::disservingrequeststats::kvcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats11kvCacheSizeE", false]], "tensorrt_llm::executor::disservingrequeststats::kvcachetransferms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats17kvCacheTransferMSE", false]], "tensorrt_llm::executor::dynamicbatchconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfigE", false]], "tensorrt_llm::executor::dynamicbatchconfig::dynamicbatchconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", false]], "tensorrt_llm::executor::dynamicbatchconfig::getbatchsizetable (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig17getBatchSizeTableEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getdynamicbatchmovingaveragewindow (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig34getDynamicBatchMovingAverageWindowEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getenablebatchsizetuning (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig24getEnableBatchSizeTuningEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getenablemaxnumtokenstuning (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig27getEnableMaxNumTokensTuningEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::kdefaultbatchsizetable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22kDefaultBatchSizeTableE", false]], "tensorrt_llm::executor::dynamicbatchconfig::kdefaultdynamicbatchmovingaveragewindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig39kDefaultDynamicBatchMovingAverageWindowE", false]], "tensorrt_llm::executor::dynamicbatchconfig::mbatchsizetable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig15mBatchSizeTableE", false]], "tensorrt_llm::executor::dynamicbatchconfig::mdynamicbatchmovingaveragewindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig32mDynamicBatchMovingAverageWindowE", false]], "tensorrt_llm::executor::dynamicbatchconfig::menablebatchsizetuning (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22mEnableBatchSizeTuningE", false]], "tensorrt_llm::executor::dynamicbatchconfig::menablemaxnumtokenstuning (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig25mEnableMaxNumTokensTuningE", false]], "tensorrt_llm::executor::eaglechoices (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12EagleChoicesE", false]], "tensorrt_llm::executor::eagleconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfigE", false]], "tensorrt_llm::executor::eagleconfig::checkposteriorvalue (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", false]], "tensorrt_llm::executor::eagleconfig::eagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::eagleconfig::getdynamictreemaxtopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getDynamicTreeMaxTopKEv", false]], "tensorrt_llm::executor::eagleconfig::geteaglechoices (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig15getEagleChoicesEv", false]], "tensorrt_llm::executor::eagleconfig::getposteriorthreshold (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getPosteriorThresholdEv", false]], "tensorrt_llm::executor::eagleconfig::isgreedysampling (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig16isGreedySamplingEv", false]], "tensorrt_llm::executor::eagleconfig::mdynamictreemaxtopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mDynamicTreeMaxTopKE", false]], "tensorrt_llm::executor::eagleconfig::meaglechoices (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig13mEagleChoicesE", false]], "tensorrt_llm::executor::eagleconfig::mgreedysampling (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mGreedySamplingE", false]], "tensorrt_llm::executor::eagleconfig::mposteriorthreshold (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mPosteriorThresholdE", false]], "tensorrt_llm::executor::eagleconfig::musedynamictree (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mUseDynamicTreeE", false]], "tensorrt_llm::executor::eagleconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", false]], "tensorrt_llm::executor::eagleconfig::usedynamictree (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig14useDynamicTreeEv", false]], "tensorrt_llm::executor::executor (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutorE", false]], "tensorrt_llm::executor::executor::awaitresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::executor::cancelrequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", false]], "tensorrt_llm::executor::executor::canenqueuerequests (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor18canEnqueueRequestsEv", false]], "tensorrt_llm::executor::executor::enqueuerequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", false]], "tensorrt_llm::executor::executor::enqueuerequests (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", false]], "tensorrt_llm::executor::executor::executor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERR8Executor", false]], "tensorrt_llm::executor::executor::getkvcacheeventmanager (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor22getKVCacheEventManagerEv", false]], "tensorrt_llm::executor::executor::getlatestdebugtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestDebugTensorsEv", false]], "tensorrt_llm::executor::executor::getlatestiterationstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor23getLatestIterationStatsEv", false]], "tensorrt_llm::executor::executor::getlatestrequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestRequestStatsEv", false]], "tensorrt_llm::executor::executor::getnumresponsesready (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", false]], "tensorrt_llm::executor::executor::isparticipant (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor13isParticipantEv", false]], "tensorrt_llm::executor::executor::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor5mImplE", false]], "tensorrt_llm::executor::executor::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", false], [0, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERR8Executor", false]], "tensorrt_llm::executor::executor::shutdown (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor8shutdownEv", false]], "tensorrt_llm::executor::executor::~executor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutorD0Ev", false]], "tensorrt_llm::executor::executorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfigE", false]], "tensorrt_llm::executor::executorconfig::executorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", false]], "tensorrt_llm::executor::executorconfig::getadditionalmodeloutputs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getAdditionalModelOutputsEv", false]], "tensorrt_llm::executor::executorconfig::getbatchingtype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getBatchingTypeEv", false]], "tensorrt_llm::executor::executorconfig::getcachetransceiverconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getCacheTransceiverConfigEv", false]], "tensorrt_llm::executor::executorconfig::getdebugconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig14getDebugConfigEv", false]], "tensorrt_llm::executor::executorconfig::getdecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getDecodingConfigEv", false]], "tensorrt_llm::executor::executorconfig::getenablechunkedcontext (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getEnableChunkedContextEv", false]], "tensorrt_llm::executor::executorconfig::getenabletrtoverlap (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getEnableTrtOverlapEv", false]], "tensorrt_llm::executor::executorconfig::getextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig32getExtendedRuntimePerfKnobConfigEv", false]], "tensorrt_llm::executor::executorconfig::getgathergenerationlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getGatherGenerationLogitsEv", false]], "tensorrt_llm::executor::executorconfig::getgpuweightspercent (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getGpuWeightsPercentEv", false]], "tensorrt_llm::executor::executorconfig::getguideddecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getGuidedDecodingConfigEv", false]], "tensorrt_llm::executor::executorconfig::getiterstatsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getIterStatsMaxIterationsEv", false]], "tensorrt_llm::executor::executorconfig::getkvcacheconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getKvCacheConfigEv", false]], "tensorrt_llm::executor::executorconfig::getkvcacheconfigref (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19getKvCacheConfigRefEv", false]], "tensorrt_llm::executor::executorconfig::getlogitspostprocessorconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getLogitsPostProcessorConfigEv", false]], "tensorrt_llm::executor::executorconfig::getmaxbatchsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBatchSizeEv", false]], "tensorrt_llm::executor::executorconfig::getmaxbeamwidth (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::executor::executorconfig::getmaxnumtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxNumTokensEv", false]], "tensorrt_llm::executor::executorconfig::getmaxqueuesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxQueueSizeEv", false]], "tensorrt_llm::executor::executorconfig::getmaxseqidlemicroseconds (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getMaxSeqIdleMicrosecondsEv", false]], "tensorrt_llm::executor::executorconfig::getnormalizelogprobs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getNormalizeLogProbsEv", false]], "tensorrt_llm::executor::executorconfig::getparallelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getParallelConfigEv", false]], "tensorrt_llm::executor::executorconfig::getpeftcacheconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getPeftCacheConfigEv", false]], "tensorrt_llm::executor::executorconfig::getprompttableoffloading (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig24getPromptTableOffloadingEv", false]], "tensorrt_llm::executor::executorconfig::getrecvpollperiodms (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getRecvPollPeriodMsEv", false]], "tensorrt_llm::executor::executorconfig::getrequeststatsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getRequestStatsMaxIterationsEv", false]], "tensorrt_llm::executor::executorconfig::getschedulerconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getSchedulerConfigEv", false]], "tensorrt_llm::executor::executorconfig::getschedulerconfigref (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21getSchedulerConfigRefEv", false]], "tensorrt_llm::executor::executorconfig::getspecdecconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getSpecDecConfigEv", false]], "tensorrt_llm::executor::executorconfig::getusegpudirectstorage (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig22getUseGpuDirectStorageEv", false]], "tensorrt_llm::executor::executorconfig::kdefaultiterstatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultIterStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::kdefaultmaxseqidlemicroseconds (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultMaxSeqIdleMicrosecondsE", false]], "tensorrt_llm::executor::executorconfig::kdefaultrequeststatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig33kDefaultRequestStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::madditionalmodeloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mAdditionalModelOutputsE", false]], "tensorrt_llm::executor::executorconfig::mbatchingtype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mBatchingTypeE", false]], "tensorrt_llm::executor::executorconfig::mcachetransceiverconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mCacheTransceiverConfigE", false]], "tensorrt_llm::executor::executorconfig::mdebugconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig12mDebugConfigE", false]], "tensorrt_llm::executor::executorconfig::mdecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::menablechunkedcontext (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mEnableChunkedContextE", false]], "tensorrt_llm::executor::executorconfig::menabletrtoverlap (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mEnableTrtOverlapE", false]], "tensorrt_llm::executor::executorconfig::mextendedruntimeperfknobconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30mExtendedRuntimePerfKnobConfigE", false]], "tensorrt_llm::executor::executorconfig::mgathergenerationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mGatherGenerationLogitsE", false]], "tensorrt_llm::executor::executorconfig::mgpuweightspercent (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mGpuWeightsPercentE", false]], "tensorrt_llm::executor::executorconfig::mguideddecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mGuidedDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::miterstatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mIterStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::mkvcacheconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14mKvCacheConfigE", false]], "tensorrt_llm::executor::executorconfig::mlogitspostprocessorconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mLogitsPostProcessorConfigE", false]], "tensorrt_llm::executor::executorconfig::mmaxbatchsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBatchSizeE", false]], "tensorrt_llm::executor::executorconfig::mmaxbeamwidth (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBeamWidthE", false]], "tensorrt_llm::executor::executorconfig::mmaxnumtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxNumTokensE", false]], "tensorrt_llm::executor::executorconfig::mmaxqueuesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxQueueSizeE", false]], "tensorrt_llm::executor::executorconfig::mmaxseqidlemicroseconds (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mMaxSeqIdleMicrosecondsE", false]], "tensorrt_llm::executor::executorconfig::mnormalizelogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mNormalizeLogProbsE", false]], "tensorrt_llm::executor::executorconfig::mparallelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mParallelConfigE", false]], "tensorrt_llm::executor::executorconfig::mpeftcacheconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mPeftCacheConfigE", false]], "tensorrt_llm::executor::executorconfig::mprompttableoffloading (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22mPromptTableOffloadingE", false]], "tensorrt_llm::executor::executorconfig::mrecvpollperiodms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mRecvPollPeriodMsE", false]], "tensorrt_llm::executor::executorconfig::mrequeststatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mRequestStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::mschedulerconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mSchedulerConfigE", false]], "tensorrt_llm::executor::executorconfig::mspeculativedecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mSpeculativeDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::musegpudirectstorage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20mUseGpuDirectStorageE", false]], "tensorrt_llm::executor::executorconfig::setadditionalmodeloutputs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", false]], "tensorrt_llm::executor::executorconfig::setbatchingtype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", false]], "tensorrt_llm::executor::executorconfig::setcachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", false]], "tensorrt_llm::executor::executorconfig::setdebugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", false]], "tensorrt_llm::executor::executorconfig::setdecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setenablechunkedcontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", false]], "tensorrt_llm::executor::executorconfig::setenabletrtoverlap (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", false]], "tensorrt_llm::executor::executorconfig::setextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", false]], "tensorrt_llm::executor::executorconfig::setgathergenerationlogits (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", false]], "tensorrt_llm::executor::executorconfig::setgpuweightspercent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", false]], "tensorrt_llm::executor::executorconfig::setguideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setiterstatsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setkvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", false]], "tensorrt_llm::executor::executorconfig::setlogitspostprocessorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", false]], "tensorrt_llm::executor::executorconfig::setmaxbatchsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxnumtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxqueuesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::executorconfig::setmaxseqidlemicroseconds (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", false]], "tensorrt_llm::executor::executorconfig::setnormalizelogprobs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", false]], "tensorrt_llm::executor::executorconfig::setparallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", false]], "tensorrt_llm::executor::executorconfig::setpeftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", false]], "tensorrt_llm::executor::executorconfig::setprompttableoffloading (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", false]], "tensorrt_llm::executor::executorconfig::setrecvpollperiodms (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setrequeststatsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setschedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", false]], "tensorrt_llm::executor::executorconfig::setspecdecconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setusegpudirectstorage (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::extendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getcudagraphcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21getCudaGraphCacheSizeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getcudagraphmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16getCudaGraphModeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getenablecontextfmhafp32acc (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27getEnableContextFMHAFP32AccEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getmultiblockmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17getMultiBlockModeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mcudagraphcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig19mCudaGraphCacheSizeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mcudagraphmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig14mCudaGraphModeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::menablecontextfmhafp32acc (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig25mEnableContextFMHAFP32AccE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mmultiblockmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig15mMultiBlockModeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setcudagraphcachesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setcudagraphmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setenablecontextfmhafp32acc (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setmultiblockmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", false]], "tensorrt_llm::executor::externaldrafttokensconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfigE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::externaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getacceptancethreshold (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig22getAcceptanceThresholdEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getfastlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig13getFastLogitsEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getLogitsEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::gettokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getTokensEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::macceptancethreshold (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig20mAcceptanceThresholdE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mfastlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig11mFastLogitsE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mLogitsE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mTokensE", false]], "tensorrt_llm::executor::finishreason (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReasonE", false]], "tensorrt_llm::executor::finishreason::kcancelled (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason10kCANCELLEDE", false]], "tensorrt_llm::executor::finishreason::kend_id (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason7kEND_IDE", false]], "tensorrt_llm::executor::finishreason::klength (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason7kLENGTHE", false]], "tensorrt_llm::executor::finishreason::knot_finished (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason13kNOT_FINISHEDE", false]], "tensorrt_llm::executor::finishreason::kstop_words (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason11kSTOP_WORDSE", false]], "tensorrt_llm::executor::finishreason::ktimed_out (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason10kTIMED_OUTE", false]], "tensorrt_llm::executor::floattype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9FloatTypeE", false]], "tensorrt_llm::executor::guideddecodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfigE", false]], "tensorrt_llm::executor::guideddecodingconfig::getbackend (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig10getBackendEv", false]], "tensorrt_llm::executor::guideddecodingconfig::getencodedvocab (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getEncodedVocabEv", false]], "tensorrt_llm::executor::guideddecodingconfig::getstoptokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getStopTokenIdsEv", false]], "tensorrt_llm::executor::guideddecodingconfig::gettokenizerstr (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getTokenizerStrEv", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingbackend (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackendE", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingbackend::kxgrammar (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackend9kXGRAMMARE", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", false]], "tensorrt_llm::executor::guideddecodingconfig::mbackend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig8mBackendE", false]], "tensorrt_llm::executor::guideddecodingconfig::mencodedvocab (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mEncodedVocabE", false]], "tensorrt_llm::executor::guideddecodingconfig::mstoptokenids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mStopTokenIdsE", false]], "tensorrt_llm::executor::guideddecodingconfig::mtokenizerstr (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mTokenizerStrE", false]], "tensorrt_llm::executor::guideddecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", false]], "tensorrt_llm::executor::guideddecodingconfig::setbackend (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", false]], "tensorrt_llm::executor::guideddecodingconfig::setencodedvocab (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", false]], "tensorrt_llm::executor::guideddecodingconfig::setstoptokenids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", false]], "tensorrt_llm::executor::guideddecodingconfig::settokenizerstr (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", false]], "tensorrt_llm::executor::guideddecodingconfig::validate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig8validateEv", false]], "tensorrt_llm::executor::guideddecodingparams (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParamsE", false]], "tensorrt_llm::executor::guideddecodingparams::getguide (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams8getGuideEv", false]], "tensorrt_llm::executor::guideddecodingparams::getguidetype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams12getGuideTypeEv", false]], "tensorrt_llm::executor::guideddecodingparams::guideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideTypeE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kebnf_grammar (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType13kEBNF_GRAMMARE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kjson (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType5kJSONE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kjson_schema (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType12kJSON_SCHEMAE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kregex (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType6kREGEXE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kstructural_tag (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType15kSTRUCTURAL_TAGE", false]], "tensorrt_llm::executor::guideddecodingparams::mguide (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams6mGuideE", false]], "tensorrt_llm::executor::guideddecodingparams::mguidetype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams10mGuideTypeE", false]], "tensorrt_llm::executor::guideddecodingparams::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", false]], "tensorrt_llm::executor::idtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6IdTypeE", false]], "tensorrt_llm::executor::inflightbatchingstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStatsE", false]], "tensorrt_llm::executor::inflightbatchingstats::avgnumdecodedtokensperiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats26avgNumDecodedTokensPerIterE", false]], "tensorrt_llm::executor::inflightbatchingstats::microbatchid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12microBatchIdE", false]], "tensorrt_llm::executor::inflightbatchingstats::numcontextrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats18numContextRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numctxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12numCtxTokensE", false]], "tensorrt_llm::executor::inflightbatchingstats::numgenrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats14numGenRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numpausedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats17numPausedRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numscheduledrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats20numScheduledRequestsE", false]], "tensorrt_llm::executor::iterationstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStatsE", false]], "tensorrt_llm::executor::iterationstats::cpumemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats11cpuMemUsageE", false]], "tensorrt_llm::executor::iterationstats::crosskvcachestats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17crossKvCacheStatsE", false]], "tensorrt_llm::executor::iterationstats::gpumemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats11gpuMemUsageE", false]], "tensorrt_llm::executor::iterationstats::inflightbatchingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats21inflightBatchingStatsE", false]], "tensorrt_llm::executor::iterationstats::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats4iterE", false]], "tensorrt_llm::executor::iterationstats::iterlatencyms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats13iterLatencyMSE", false]], "tensorrt_llm::executor::iterationstats::kvcachestats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats12kvCacheStatsE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizeruntime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxBatchSizeRuntimeE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizestatic (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxBatchSizeStaticE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizetunerrecommended (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxBatchSizeTunerRecommendedE", false]], "tensorrt_llm::executor::iterationstats::maxnumactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20maxNumActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokensruntime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxNumTokensRuntimeE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokensstatic (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxNumTokensStaticE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokenstunerrecommended (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxNumTokensTunerRecommendedE", false]], "tensorrt_llm::executor::iterationstats::newactiverequestsqueuelatencyms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats31newActiveRequestsQueueLatencyMSE", false]], "tensorrt_llm::executor::iterationstats::numactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17numActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::numcompletedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20numCompletedRequestsE", false]], "tensorrt_llm::executor::iterationstats::numnewactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20numNewActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::numqueuedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17numQueuedRequestsE", false]], "tensorrt_llm::executor::iterationstats::pinnedmemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats14pinnedMemUsageE", false]], "tensorrt_llm::executor::iterationstats::staticbatchingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19staticBatchingStatsE", false]], "tensorrt_llm::executor::iterationstats::timestamp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats9timestampE", false]], "tensorrt_llm::executor::iterationtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor13IterationTypeE", false]], "tensorrt_llm::executor::jsonserialization (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor17JsonSerializationE", false]], "tensorrt_llm::executor::jsonserialization::tojsonstr (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", false]], "tensorrt_llm::executor::kv_cache (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", false]], "tensorrt_llm::executor::kv_cache::cachestate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::attentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::mattentiontype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig14mAttentionTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::mkvfactor (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig9mKvFactorE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype::kdefault (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType8kDEFAULTE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype::kmla (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType4kMLAE", false]], "tensorrt_llm::executor::kv_cache::cachestate::cachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", false]], "tensorrt_llm::executor::kv_cache::cachestate::getattentionconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState18getAttentionConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getdatatype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11getDataTypeEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getmodelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14getModelConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getparallelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState17getParallelConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::mattentionconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState16mAttentionConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::mdatatype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState9mDataTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::mmodelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState12mModelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::mnbkvheadsperlayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig18mNbKvHeadsPerLayerE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::msizeperhead (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig12mSizePerHeadE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::mtokensperblock (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig15mTokensPerBlockE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", false]], "tensorrt_llm::executor::kv_cache::cachestate::mparallelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15mParallelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mdprank (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPrankE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mdpsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPsizeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::menableattentiondp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mEnableAttentionDPE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mpipelineparallelism (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig20mPipelineParallelismE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mtensorparallelism (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mTensorParallelismE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", false]], "tensorrt_llm::executor::kv_cache::cachestate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::commstate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommStateE", false]], "tensorrt_llm::executor::kv_cache::commstate::commstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getmpistate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState11getMpiStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getselfidx (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10getSelfIdxEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getsocketstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState14getSocketStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::ismpistate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10isMpiStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::issocketstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13isSocketStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::mselfidx (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState8mSelfIdxE", false]], "tensorrt_llm::executor::kv_cache::commstate::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState6mStateE", false]], "tensorrt_llm::executor::kv_cache::commstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", false]], "tensorrt_llm::executor::kv_cache::commstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::connection (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionE", false]], "tensorrt_llm::executor::kv_cache::connection::isthreadsafe (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection12isThreadSafeEv", false]], "tensorrt_llm::executor::kv_cache::connection::recv (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", false]], "tensorrt_llm::executor::kv_cache::connection::send (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", false]], "tensorrt_llm::executor::kv_cache::connection::~connection (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionD0Ev", false]], "tensorrt_llm::executor::kv_cache::connectionmanager (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerE", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::getcommstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache17ConnectionManager12getCommStateEv", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::getconnections (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::recvconnect (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::~connectionmanager (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerD0Ev", false]], "tensorrt_llm::executor::kv_cache::datacontext (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContextE", false]], "tensorrt_llm::executor::kv_cache::datacontext::datacontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", false]], "tensorrt_llm::executor::kv_cache::datacontext::gettag (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11DataContext6getTagEv", false]], "tensorrt_llm::executor::kv_cache::datacontext::mtag (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext4mTagE", false]], "tensorrt_llm::executor::kv_cache::mpistate (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiStateE", false]], "tensorrt_llm::executor::kv_cache::mpistate::mranks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiState6mRanksE", false]], "tensorrt_llm::executor::kv_cache::mpistate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", false]], "tensorrt_llm::executor::kv_cache::mpistate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::socketstate (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketStateE", false]], "tensorrt_llm::executor::kv_cache::socketstate::mip (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState3mIpE", false]], "tensorrt_llm::executor::kv_cache::socketstate::mport (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState5mPortE", false]], "tensorrt_llm::executor::kv_cache::socketstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", false]], "tensorrt_llm::executor::kv_cache::socketstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketState8toStringEv", false]], "tensorrt_llm::executor::kvcacheconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfigE", false]], "tensorrt_llm::executor::kvcacheconfig::fillemptyfieldsfromruntimedefaults (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", false]], "tensorrt_llm::executor::kvcacheconfig::getcopyonpartialreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getCopyOnPartialReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::getcrosskvcachefraction (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig23getCrossKvCacheFractionEv", false]], "tensorrt_llm::executor::kvcacheconfig::getenableblockreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig19getEnableBlockReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::getenablepartialreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEnablePartialReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::geteventbuffermaxsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEventBufferMaxSizeEv", false]], "tensorrt_llm::executor::kvcacheconfig::getfreegpumemoryfraction (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getFreeGpuMemoryFractionEv", false]], "tensorrt_llm::executor::kvcacheconfig::gethostcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getHostCacheSizeEv", false]], "tensorrt_llm::executor::kvcacheconfig::getmaxattentionwindowvec (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getMaxAttentionWindowVecEv", false]], "tensorrt_llm::executor::kvcacheconfig::getmaxtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig12getMaxTokensEv", false]], "tensorrt_llm::executor::kvcacheconfig::getonboardblocks (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getOnboardBlocksEv", false]], "tensorrt_llm::executor::kvcacheconfig::getsecondaryoffloadminpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig30getSecondaryOffloadMinPriorityEv", false]], "tensorrt_llm::executor::kvcacheconfig::getsinktokenlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig18getSinkTokenLengthEv", false]], "tensorrt_llm::executor::kvcacheconfig::kvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", false]], "tensorrt_llm::executor::kvcacheconfig::mcopyonpartialreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mCopyOnPartialReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::mcrosskvcachefraction (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21mCrossKvCacheFractionE", false]], "tensorrt_llm::executor::kvcacheconfig::menableblockreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig17mEnableBlockReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::menablepartialreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEnablePartialReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::meventbuffermaxsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEventBufferMaxSizeE", false]], "tensorrt_llm::executor::kvcacheconfig::mfreegpumemoryfraction (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mFreeGpuMemoryFractionE", false]], "tensorrt_llm::executor::kvcacheconfig::mhostcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mHostCacheSizeE", false]], "tensorrt_llm::executor::kvcacheconfig::mmaxattentionwindowvec (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mMaxAttentionWindowVecE", false]], "tensorrt_llm::executor::kvcacheconfig::mmaxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig10mMaxTokensE", false]], "tensorrt_llm::executor::kvcacheconfig::monboardblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mOnboardBlocksE", false]], "tensorrt_llm::executor::kvcacheconfig::msecondaryoffloadminpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig28mSecondaryOffloadMinPriorityE", false]], "tensorrt_llm::executor::kvcacheconfig::msinktokenlength (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16mSinkTokenLengthE", false]], "tensorrt_llm::executor::kvcacheconfig::setcopyonpartialreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::setcrosskvcachefraction (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", false]], "tensorrt_llm::executor::kvcacheconfig::setenableblockreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::setenablepartialreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::seteventbuffermaxsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", false]], "tensorrt_llm::executor::kvcacheconfig::setfreegpumemoryfraction (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", false]], "tensorrt_llm::executor::kvcacheconfig::sethostcachesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", false]], "tensorrt_llm::executor::kvcacheconfig::setmaxattentionwindowvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::kvcacheconfig::setmaxtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", false]], "tensorrt_llm::executor::kvcacheconfig::setonboardblocks (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", false]], "tensorrt_llm::executor::kvcacheconfig::setsecondaryoffloadminpriority (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", false]], "tensorrt_llm::executor::kvcacheconfig::setsinktokenlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", false]], "tensorrt_llm::executor::kvcachecreateddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedDataE", false]], "tensorrt_llm::executor::kvcachecreateddata::numblockspercachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedData22numBlocksPerCacheLevelE", false]], "tensorrt_llm::executor::kvcacheevent (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEventE", false]], "tensorrt_llm::executor::kvcacheevent::data (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent4dataE", false]], "tensorrt_llm::executor::kvcacheevent::eventid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent7eventIdE", false]], "tensorrt_llm::executor::kvcacheevent::kvcacheevent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", false]], "tensorrt_llm::executor::kvcacheeventdata (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDataE", false]], "tensorrt_llm::executor::kvcacheeventdiff (c++ struct)": [[0, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", false]], "tensorrt_llm::executor::kvcacheeventdiff::newvalue (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8newValueE", false]], "tensorrt_llm::executor::kvcacheeventdiff::oldvalue (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8oldValueE", false]], "tensorrt_llm::executor::kvcacheeventmanager (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManagerE", false]], "tensorrt_llm::executor::kvcacheeventmanager::getlatestevents (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::kvcacheeventmanager::kvcacheeventmanager (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", false]], "tensorrt_llm::executor::kvcacheeventmanager::kvcachemanager (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager14kvCacheManagerE", false]], "tensorrt_llm::executor::kvcacheremoveddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedDataE", false]], "tensorrt_llm::executor::kvcacheremoveddata::blockhashes (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedData11blockHashesE", false]], "tensorrt_llm::executor::kvcacheretentionconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfigE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getdecodedurationms (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig19getDecodeDurationMsEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getdecoderetentionpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig26getDecodeRetentionPriorityEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getperblockretentionpriorityduration (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcacheretentionconfig::gettokenrangeretentionconfigs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig29getTokenRangeRetentionConfigsEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kdefaultretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25kDefaultRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kmaxretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMaxRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kminretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMinRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", false], [0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mdecodedurationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig17mDecodeDurationMsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mdecoderetentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig24mDecodeRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mtokenrangeretentionconfigs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig27mTokenRangeRetentionConfigsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::durationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10durationMsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8priorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8tokenEndE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenrangeretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenstart (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10tokenStartE", false]], "tensorrt_llm::executor::kvcachestats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStatsE", false]], "tensorrt_llm::executor::kvcachestats::allocnewblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14allocNewBlocksE", false]], "tensorrt_llm::executor::kvcachestats::alloctotalblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats16allocTotalBlocksE", false]], "tensorrt_llm::executor::kvcachestats::cachehitrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12cacheHitRateE", false]], "tensorrt_llm::executor::kvcachestats::freenumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13freeNumBlocksE", false]], "tensorrt_llm::executor::kvcachestats::maxnumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12maxNumBlocksE", false]], "tensorrt_llm::executor::kvcachestats::missedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12missedBlocksE", false]], "tensorrt_llm::executor::kvcachestats::reusedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12reusedBlocksE", false]], "tensorrt_llm::executor::kvcachestats::tokensperblock (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14tokensPerBlockE", false]], "tensorrt_llm::executor::kvcachestats::usednumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13usedNumBlocksE", false]], "tensorrt_llm::executor::kvcachestoredblockdata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockDataE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::blockhash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData9blockHashE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::cachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData10cacheLevelE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::kvcachestoredblockdata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcachestoredblockdata::loraid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6loraIdE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData8priorityE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::tokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6tokensE", false]], "tensorrt_llm::executor::kvcachestoreddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredDataE", false]], "tensorrt_llm::executor::kvcachestoreddata::blocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData6blocksE", false]], "tensorrt_llm::executor::kvcachestoreddata::parenthash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData10parentHashE", false]], "tensorrt_llm::executor::kvcacheupdateddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedDataE", false]], "tensorrt_llm::executor::kvcacheupdateddata::blockhash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData9blockHashE", false]], "tensorrt_llm::executor::kvcacheupdateddata::cachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData10cacheLevelE", false]], "tensorrt_llm::executor::kvcacheupdateddata::cachelevelupdated (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcacheupdateddata::kvcacheupdateddata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", false]], "tensorrt_llm::executor::kvcacheupdateddata::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData8priorityE", false]], "tensorrt_llm::executor::kvcacheupdateddata::priorityupdated (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::logitspostprocessor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor19LogitsPostProcessorE", false]], "tensorrt_llm::executor::logitspostprocessorbatched (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor26LogitsPostProcessorBatchedE", false]], "tensorrt_llm::executor::logitspostprocessorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfigE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getprocessorbatched (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig19getProcessorBatchedEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getprocessormap (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig15getProcessorMapEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getreplicate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig12getReplicateEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::logitspostprocessorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mprocessorbatched (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig17mProcessorBatchedE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mprocessormap (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig13mProcessorMapE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mreplicate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig10mReplicateE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setprocessorbatched (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setprocessormap (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setreplicate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", false]], "tensorrt_llm::executor::logitspostprocessormap (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor22LogitsPostProcessorMapE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfigE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::calculatespeculativeresource (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig28calculateSpeculativeResourceEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::calculatespeculativeresourcetuple (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::get (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig3getEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig12getNgramSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getverificationsetsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig22getVerificationSetSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getwindowsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig13getWindowSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::isle (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::islegal (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingngram (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig30kDefaultLookaheadDecodingNgramE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingverificationset (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig40kDefaultLookaheadDecodingVerificationSetE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingwindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig31kDefaultLookaheadDecodingWindowE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::lookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", false], [0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig10mNgramSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mverificationsetsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig20mVerificationSetSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mwindowsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig11mWindowSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::loraconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfigE", false]], "tensorrt_llm::executor::loraconfig::getconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getConfigEv", false]], "tensorrt_llm::executor::loraconfig::gettaskid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getTaskIdEv", false]], "tensorrt_llm::executor::loraconfig::getweights (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig10getWeightsEv", false]], "tensorrt_llm::executor::loraconfig::loraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", false]], "tensorrt_llm::executor::loraconfig::mconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mConfigE", false]], "tensorrt_llm::executor::loraconfig::mtaskid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mTaskIdE", false]], "tensorrt_llm::executor::loraconfig::mweights (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig8mWeightsE", false]], "tensorrt_llm::executor::medusachoices (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor13MedusaChoicesE", false]], "tensorrt_llm::executor::memorytype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryTypeE", false]], "tensorrt_llm::executor::memorytype::kcpu (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kCPUE", false]], "tensorrt_llm::executor::memorytype::kcpu_pinned (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType11kCPU_PINNEDE", false]], "tensorrt_llm::executor::memorytype::kcpu_pinnedpool (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType15kCPU_PINNEDPOOLE", false]], "tensorrt_llm::executor::memorytype::kgpu (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kGPUE", false]], "tensorrt_llm::executor::memorytype::kunknown (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType8kUNKNOWNE", false]], "tensorrt_llm::executor::memorytype::kuvm (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kUVME", false]], "tensorrt_llm::executor::millisecondstype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16MillisecondsTypeE", false]], "tensorrt_llm::executor::modeltype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelTypeE", false]], "tensorrt_llm::executor::modeltype::kdecoder_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType13kDECODER_ONLYE", false]], "tensorrt_llm::executor::modeltype::kencoder_decoder (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType16kENCODER_DECODERE", false]], "tensorrt_llm::executor::modeltype::kencoder_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType13kENCODER_ONLYE", false]], "tensorrt_llm::executor::mropeconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfigE", false]], "tensorrt_llm::executor::mropeconfig::getmropepositiondeltas (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11MropeConfig22getMRopePositionDeltasEv", false]], "tensorrt_llm::executor::mropeconfig::getmroperotarycossin (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11MropeConfig20getMRopeRotaryCosSinEv", false]], "tensorrt_llm::executor::mropeconfig::mmropepositiondeltas (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig20mMRopePositionDeltasE", false]], "tensorrt_llm::executor::mropeconfig::mmroperotarycossin (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig18mMRopeRotaryCosSinE", false]], "tensorrt_llm::executor::mropeconfig::mropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", false]], "tensorrt_llm::executor::operator<< (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", false], [0, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", false]], "tensorrt_llm::executor::orchestratorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfigE", false]], "tensorrt_llm::executor::orchestratorconfig::getisorchestrator (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getIsOrchestratorEv", false]], "tensorrt_llm::executor::orchestratorconfig::getorchleadercomm (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getOrchLeaderCommEv", false]], "tensorrt_llm::executor::orchestratorconfig::getspawnprocesses (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getSpawnProcessesEv", false]], "tensorrt_llm::executor::orchestratorconfig::getworkerexecutablepath (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig23getWorkerExecutablePathEv", false]], "tensorrt_llm::executor::orchestratorconfig::misorchestrator (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mIsOrchestratorE", false]], "tensorrt_llm::executor::orchestratorconfig::morchleadercomm (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mOrchLeaderCommE", false]], "tensorrt_llm::executor::orchestratorconfig::mspawnprocesses (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mSpawnProcessesE", false]], "tensorrt_llm::executor::orchestratorconfig::mworkerexecutablepath (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig21mWorkerExecutablePathE", false]], "tensorrt_llm::executor::orchestratorconfig::orchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", false]], "tensorrt_llm::executor::orchestratorconfig::setisorchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", false]], "tensorrt_llm::executor::orchestratorconfig::setorchleadercomm (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", false]], "tensorrt_llm::executor::orchestratorconfig::setspawnprocesses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", false]], "tensorrt_llm::executor::orchestratorconfig::setworkerexecutablepath (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", false]], "tensorrt_llm::executor::outputconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfigE", false]], "tensorrt_llm::executor::outputconfig::additionalmodeloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22additionalModelOutputsE", false]], "tensorrt_llm::executor::outputconfig::excludeinputfromoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22excludeInputFromOutputE", false]], "tensorrt_llm::executor::outputconfig::outputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", false]], "tensorrt_llm::executor::outputconfig::returncontextlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnContextLogitsE", false]], "tensorrt_llm::executor::outputconfig::returnencoderoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnEncoderOutputE", false]], "tensorrt_llm::executor::outputconfig::returngenerationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22returnGenerationLogitsE", false]], "tensorrt_llm::executor::outputconfig::returnlogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig14returnLogProbsE", false]], "tensorrt_llm::executor::outputconfig::returnperfmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig17returnPerfMetricsE", false]], "tensorrt_llm::executor::parallelconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfigE", false]], "tensorrt_llm::executor::parallelconfig::getcommunicationmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationModeEv", false]], "tensorrt_llm::executor::parallelconfig::getcommunicationtype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationTypeEv", false]], "tensorrt_llm::executor::parallelconfig::getdeviceids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig12getDeviceIdsEv", false]], "tensorrt_llm::executor::parallelconfig::getnumnodes (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig11getNumNodesEv", false]], "tensorrt_llm::executor::parallelconfig::getorchestratorconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig21getOrchestratorConfigEv", false]], "tensorrt_llm::executor::parallelconfig::getparticipantids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig17getParticipantIdsEv", false]], "tensorrt_llm::executor::parallelconfig::mcommmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommModeE", false]], "tensorrt_llm::executor::parallelconfig::mcommtype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommTypeE", false]], "tensorrt_llm::executor::parallelconfig::mdeviceids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig10mDeviceIdsE", false]], "tensorrt_llm::executor::parallelconfig::mnumnodes (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mNumNodesE", false]], "tensorrt_llm::executor::parallelconfig::morchestratorconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig19mOrchestratorConfigE", false]], "tensorrt_llm::executor::parallelconfig::mparticipantids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig15mParticipantIdsE", false]], "tensorrt_llm::executor::parallelconfig::parallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::parallelconfig::setcommunicationmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", false]], "tensorrt_llm::executor::parallelconfig::setcommunicationtype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", false]], "tensorrt_llm::executor::parallelconfig::setdeviceids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::parallelconfig::setnumnodes (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", false]], "tensorrt_llm::executor::parallelconfig::setorchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", false]], "tensorrt_llm::executor::parallelconfig::setparticipantids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::peftcacheconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfigE", false]], "tensorrt_llm::executor::peftcacheconfig::getdevicecachepercent (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getDeviceCachePercentEv", false]], "tensorrt_llm::executor::peftcacheconfig::gethostcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getHostCacheSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::getloraprefetchdir (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig18getLoraPrefetchDirEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxadaptersize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getMaxAdapterSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxpagesperblockdevice (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig25getMaxPagesPerBlockDeviceEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxpagesperblockhost (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getMaxPagesPerBlockHostEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumcopystreams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getNumCopyStreamsEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumdevicemodulelayer (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getNumDeviceModuleLayerEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumensureworkers (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig19getNumEnsureWorkersEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumhostmodulelayer (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getNumHostModuleLayerEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumputworkers (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getNumPutWorkersEv", false]], "tensorrt_llm::executor::peftcacheconfig::getoptimaladaptersize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getOptimalAdapterSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxadaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig22kDefaultMaxAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxpagesperblockdevice (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig30kDefaultMaxPagesPerBlockDeviceE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxpagesperblockhost (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig28kDefaultMaxPagesPerBlockHostE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultoptimaladaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig26kDefaultOptimalAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mdevicecachepercent (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mDeviceCachePercentE", false]], "tensorrt_llm::executor::peftcacheconfig::mhostcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mHostCacheSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mloraprefetchdir (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig16mLoraPrefetchDirE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxadaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mMaxAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxpagesperblockdevice (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig23mMaxPagesPerBlockDeviceE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxpagesperblockhost (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mMaxPagesPerBlockHostE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumcopystreams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mNumCopyStreamsE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumdevicemodulelayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mNumDeviceModuleLayerE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumensureworkers (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig17mNumEnsureWorkersE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumhostmodulelayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mNumHostModuleLayerE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumputworkers (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mNumPutWorkersE", false]], "tensorrt_llm::executor::peftcacheconfig::moptimaladaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mOptimalAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", false]], "tensorrt_llm::executor::peftcacheconfig::peftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", false]], "tensorrt_llm::executor::prioritytype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12PriorityTypeE", false]], "tensorrt_llm::executor::prompttuningconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfigE", false]], "tensorrt_llm::executor::prompttuningconfig::getembeddingtable (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig17getEmbeddingTableEv", false]], "tensorrt_llm::executor::prompttuningconfig::getinputtokenextraids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig21getInputTokenExtraIdsEv", false]], "tensorrt_llm::executor::prompttuningconfig::membeddingtable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig15mEmbeddingTableE", false]], "tensorrt_llm::executor::prompttuningconfig::minputtokenextraids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig19mInputTokenExtraIdsE", false]], "tensorrt_llm::executor::prompttuningconfig::prompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", false]], "tensorrt_llm::executor::randomseedtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor14RandomSeedTypeE", false]], "tensorrt_llm::executor::request (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestE", false]], "tensorrt_llm::executor::request::getadditionaloutputnames (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request24getAdditionalOutputNamesEv", false]], "tensorrt_llm::executor::request::getallottedtimems (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request17getAllottedTimeMsEv", false]], "tensorrt_llm::executor::request::getbadwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getBadWordsEv", false]], "tensorrt_llm::executor::request::getclientid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getClientIdEv", false]], "tensorrt_llm::executor::request::getcontextphaseparams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getContextPhaseParamsEv", false]], "tensorrt_llm::executor::request::getcrossattentionmask (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getCrossAttentionMaskEv", false]], "tensorrt_llm::executor::request::geteagleconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getEagleConfigEv", false]], "tensorrt_llm::executor::request::getembeddingbias (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request16getEmbeddingBiasEv", false]], "tensorrt_llm::executor::request::getencoderinputfeatures (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputFeaturesEv", false]], "tensorrt_llm::executor::request::getencoderinputtokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputTokenIdsEv", false]], "tensorrt_llm::executor::request::getencoderoutputlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getEncoderOutputLengthEv", false]], "tensorrt_llm::executor::request::getendid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request8getEndIdEv", false]], "tensorrt_llm::executor::request::getexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request28getExternalDraftTokensConfigEv", false]], "tensorrt_llm::executor::request::getguideddecodingparams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getGuidedDecodingParamsEv", false]], "tensorrt_llm::executor::request::getinputtokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request16getInputTokenIdsEv", false]], "tensorrt_llm::executor::request::getkvcacheretentionconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request25getKvCacheRetentionConfigEv", false]], "tensorrt_llm::executor::request::getlanguageadapteruid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getLanguageAdapterUidEv", false]], "tensorrt_llm::executor::request::getlogitspostprocessor (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getLogitsPostProcessorEv", false]], "tensorrt_llm::executor::request::getlogitspostprocessorname (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request26getLogitsPostProcessorNameEv", false]], "tensorrt_llm::executor::request::getlookaheadconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request18getLookaheadConfigEv", false]], "tensorrt_llm::executor::request::getloraconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request13getLoraConfigEv", false]], "tensorrt_llm::executor::request::getmaxtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getMaxTokensEv", false]], "tensorrt_llm::executor::request::getmropeconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getMropeConfigEv", false]], "tensorrt_llm::executor::request::getmultimodalembedding (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getMultimodalEmbeddingEv", false]], "tensorrt_llm::executor::request::getoutputconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request15getOutputConfigEv", false]], "tensorrt_llm::executor::request::getpadid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request8getPadIdEv", false]], "tensorrt_llm::executor::request::getpositionids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getPositionIdsEv", false]], "tensorrt_llm::executor::request::getpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getPriorityEv", false]], "tensorrt_llm::executor::request::getprompttuningconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getPromptTuningConfigEv", false]], "tensorrt_llm::executor::request::getrequesttype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getRequestTypeEv", false]], "tensorrt_llm::executor::request::getreturnallgeneratedtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request27getReturnAllGeneratedTokensEv", false]], "tensorrt_llm::executor::request::getsamplingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request17getSamplingConfigEv", false]], "tensorrt_llm::executor::request::getskipcrossattnblocks (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getSkipCrossAttnBlocksEv", false]], "tensorrt_llm::executor::request::getstopwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getStopWordsEv", false]], "tensorrt_llm::executor::request::getstreaming (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getStreamingEv", false]], "tensorrt_llm::executor::request::kbatchedpostprocessorname (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request25kBatchedPostProcessorNameE", false]], "tensorrt_llm::executor::request::kdefaultpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request16kDefaultPriorityE", false]], "tensorrt_llm::executor::request::kdynamicpostprocessornameprefix (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request31kDynamicPostProcessorNamePrefixE", false]], "tensorrt_llm::executor::request::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request5mImplE", false]], "tensorrt_llm::executor::request::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", false]], "tensorrt_llm::executor::request::request (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", false]], "tensorrt_llm::executor::request::setallottedtimems (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", false]], "tensorrt_llm::executor::request::setbadwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", false]], "tensorrt_llm::executor::request::setclientid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", false]], "tensorrt_llm::executor::request::setcontextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", false]], "tensorrt_llm::executor::request::setcrossattentionmask (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", false]], "tensorrt_llm::executor::request::seteagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", false]], "tensorrt_llm::executor::request::setembeddingbias (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", false]], "tensorrt_llm::executor::request::setencoderinputfeatures (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", false]], "tensorrt_llm::executor::request::setencoderinputtokenids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", false]], "tensorrt_llm::executor::request::setencoderoutputlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", false]], "tensorrt_llm::executor::request::setendid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", false]], "tensorrt_llm::executor::request::setexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", false]], "tensorrt_llm::executor::request::setguideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", false]], "tensorrt_llm::executor::request::setkvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", false]], "tensorrt_llm::executor::request::setlanguageadapteruid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", false]], "tensorrt_llm::executor::request::setlogitspostprocessor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", false]], "tensorrt_llm::executor::request::setlogitspostprocessorname (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", false]], "tensorrt_llm::executor::request::setlookaheadconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::request::setloraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", false]], "tensorrt_llm::executor::request::setmropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", false]], "tensorrt_llm::executor::request::setmultimodalembedding (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", false]], "tensorrt_llm::executor::request::setoutputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", false]], "tensorrt_llm::executor::request::setpadid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", false]], "tensorrt_llm::executor::request::setpositionids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::request::setpriority (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", false]], "tensorrt_llm::executor::request::setprompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", false]], "tensorrt_llm::executor::request::setrequesttype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", false]], "tensorrt_llm::executor::request::setreturnallgeneratedtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", false]], "tensorrt_llm::executor::request::setsamplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", false]], "tensorrt_llm::executor::request::setskipcrossattnblocks (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", false]], "tensorrt_llm::executor::request::setstopwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", false]], "tensorrt_llm::executor::request::setstreaming (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", false]], "tensorrt_llm::executor::request::~request (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestD0Ev", false]], "tensorrt_llm::executor::requestperfmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::firstiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9firstIterE", false]], "tensorrt_llm::executor::requestperfmetrics::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics4iterE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14kvCacheMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::kvcachehitrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics14kvCacheHitRateE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::nummissedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numMissedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numnewallocatedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics21numNewAllocatedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numreusedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numReusedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numtotalallocatedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics23numTotalAllocatedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::lastiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics8lastIterE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecoding (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics19speculativeDecodingE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::acceptancerate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics14acceptanceRateE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::totalaccepteddrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics24totalAcceptedDraftTokensE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::totaldrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics16totalDraftTokensE", false]], "tensorrt_llm::executor::requestperfmetrics::timepoint (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9TimePointE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13timingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::arrivaltime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11arrivalTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::firstscheduledtime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18firstScheduledTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::firsttokentime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics14firstTokenTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11kvCacheSizeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachetransferend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18kvCacheTransferEndE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachetransferstart (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics20kvCacheTransferStartE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::lasttokentime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics13lastTokenTimeE", false]], "tensorrt_llm::executor::requeststage (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStageE", false]], "tensorrt_llm::executor::requeststage::kcontext_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kCONTEXT_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kencoder_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kENCODER_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kgeneration_complete (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kGENERATION_COMPLETEE", false]], "tensorrt_llm::executor::requeststage::kgeneration_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage23kGENERATION_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kqueued (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage7kQUEUEDE", false]], "tensorrt_llm::executor::requeststats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStatsE", false]], "tensorrt_llm::executor::requeststats::allocnewblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats24allocNewBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::alloctotalblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats26allocTotalBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::avgnumdecodedtokensperiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats26avgNumDecodedTokensPerIterE", false]], "tensorrt_llm::executor::requeststats::contextprefillposition (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22contextPrefillPositionE", false]], "tensorrt_llm::executor::requeststats::disservingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats15disServingStatsE", false]], "tensorrt_llm::executor::requeststats::id (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats2idE", false]], "tensorrt_llm::executor::requeststats::kvcachehitrateperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats24kvCacheHitRatePerRequestE", false]], "tensorrt_llm::executor::requeststats::missedblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22missedBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::numgeneratedtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats18numGeneratedTokensE", false]], "tensorrt_llm::executor::requeststats::paused (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats6pausedE", false]], "tensorrt_llm::executor::requeststats::reusedblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22reusedBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::scheduled (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats9scheduledE", false]], "tensorrt_llm::executor::requeststats::stage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats5stageE", false]], "tensorrt_llm::executor::requeststatsperiteration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIterationE", false]], "tensorrt_llm::executor::requeststatsperiteration::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration4iterE", false]], "tensorrt_llm::executor::requeststatsperiteration::requeststats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration12requestStatsE", false]], "tensorrt_llm::executor::requesttype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestTypeE", false]], "tensorrt_llm::executor::requesttype::request_type_context_and_generation (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType35REQUEST_TYPE_CONTEXT_AND_GENERATIONE", false]], "tensorrt_llm::executor::requesttype::request_type_context_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType25REQUEST_TYPE_CONTEXT_ONLYE", false]], "tensorrt_llm::executor::requesttype::request_type_generation_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType28REQUEST_TYPE_GENERATION_ONLYE", false]], "tensorrt_llm::executor::response (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseE", false]], "tensorrt_llm::executor::response::getclientid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response11getClientIdEv", false]], "tensorrt_llm::executor::response::geterrormsg (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response11getErrorMsgEv", false]], "tensorrt_llm::executor::response::getrequestid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response12getRequestIdEv", false]], "tensorrt_llm::executor::response::getresult (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response9getResultEv", false]], "tensorrt_llm::executor::response::haserror (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response8hasErrorEv", false]], "tensorrt_llm::executor::response::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8Response5mImplE", false]], "tensorrt_llm::executor::response::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", false]], "tensorrt_llm::executor::response::response (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", false]], "tensorrt_llm::executor::response::~response (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseD0Ev", false]], "tensorrt_llm::executor::result (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor6ResultE", false]], "tensorrt_llm::executor::result::additionaloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result17additionalOutputsE", false]], "tensorrt_llm::executor::result::contextlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13contextLogitsE", false]], "tensorrt_llm::executor::result::contextphaseparams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result18contextPhaseParamsE", false]], "tensorrt_llm::executor::result::cumlogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result11cumLogProbsE", false]], "tensorrt_llm::executor::result::decodingiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result12decodingIterE", false]], "tensorrt_llm::executor::result::encoderoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13encoderOutputE", false]], "tensorrt_llm::executor::result::finishreasons (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13finishReasonsE", false]], "tensorrt_llm::executor::result::generationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result16generationLogitsE", false]], "tensorrt_llm::executor::result::isfinal (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result7isFinalE", false]], "tensorrt_llm::executor::result::issequencefinal (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result15isSequenceFinalE", false]], "tensorrt_llm::executor::result::logprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result8logProbsE", false]], "tensorrt_llm::executor::result::outputtokenids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result14outputTokenIdsE", false]], "tensorrt_llm::executor::result::requestperfmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result18requestPerfMetricsE", false]], "tensorrt_llm::executor::result::sequenceindex (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13sequenceIndexE", false]], "tensorrt_llm::executor::result::specdecfastlogitsinfo (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result21specDecFastLogitsInfoE", false]], "tensorrt_llm::executor::retentionpriority (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor17RetentionPriorityE", false]], "tensorrt_llm::executor::retentionpriorityandduration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDurationE", false]], "tensorrt_llm::executor::retentionpriorityandduration::durationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration10durationMsE", false]], "tensorrt_llm::executor::retentionpriorityandduration::retentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration17retentionPriorityE", false]], "tensorrt_llm::executor::retentionpriorityandduration::retentionpriorityandduration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::samplingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfigE", false]], "tensorrt_llm::executor::samplingconfig::checkbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkbeamwidtharray (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkearlystopping (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checklengthpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkmintokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checknorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checknumreturnsequences (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkrepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktopk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppdecay (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppmin (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppresetids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", false]], "tensorrt_llm::executor::samplingconfig::getbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig26getBeamSearchDiversityRateEv", false]], "tensorrt_llm::executor::samplingconfig::getbeamwidth (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getBeamWidthEv", false]], "tensorrt_llm::executor::samplingconfig::getbeamwidtharray (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getBeamWidthArrayEv", false]], "tensorrt_llm::executor::samplingconfig::getearlystopping (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getEarlyStoppingEv", false]], "tensorrt_llm::executor::samplingconfig::getfrequencypenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig19getFrequencyPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getlengthpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getLengthPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getminp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getMinPEv", false]], "tensorrt_llm::executor::samplingconfig::getmintokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getMinTokensEv", false]], "tensorrt_llm::executor::samplingconfig::getnorepeatngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getNoRepeatNgramSizeEv", false]], "tensorrt_llm::executor::samplingconfig::getnumreturnbeams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getNumReturnBeamsEv", false]], "tensorrt_llm::executor::samplingconfig::getnumreturnsequences (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig21getNumReturnSequencesEv", false]], "tensorrt_llm::executor::samplingconfig::getpresencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig18getPresencePenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getrepetitionpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getRepetitionPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getseed (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getSeedEv", false]], "tensorrt_llm::executor::samplingconfig::gettemperature (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig14getTemperatureEv", false]], "tensorrt_llm::executor::samplingconfig::gettopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopKEv", false]], "tensorrt_llm::executor::samplingconfig::gettopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopPEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppdecay (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getTopPDecayEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppmin (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig10getTopPMinEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppresetids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig15getTopPResetIdsEv", false]], "tensorrt_llm::executor::samplingconfig::mbeamsearchdiversityrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig24mBeamSearchDiversityRateE", false]], "tensorrt_llm::executor::samplingconfig::mbeamwidth (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mBeamWidthE", false]], "tensorrt_llm::executor::samplingconfig::mbeamwidtharray (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mBeamWidthArrayE", false]], "tensorrt_llm::executor::samplingconfig::mearlystopping (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mEarlyStoppingE", false]], "tensorrt_llm::executor::samplingconfig::mfrequencypenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17mFrequencyPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mlengthpenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mLengthPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mminp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mMinPE", false]], "tensorrt_llm::executor::samplingconfig::mmintokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mMinTokensE", false]], "tensorrt_llm::executor::samplingconfig::mnorepeatngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mNoRepeatNgramSizeE", false]], "tensorrt_llm::executor::samplingconfig::mnumreturnbeams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mNumReturnBeamsE", false]], "tensorrt_llm::executor::samplingconfig::mnumreturnsequences (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19mNumReturnSequencesE", false]], "tensorrt_llm::executor::samplingconfig::mpresencepenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16mPresencePenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mrepetitionpenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mRepetitionPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mseed (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mSeedE", false]], "tensorrt_llm::executor::samplingconfig::mtemperature (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12mTemperatureE", false]], "tensorrt_llm::executor::samplingconfig::mtopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopKE", false]], "tensorrt_llm::executor::samplingconfig::mtopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopPE", false]], "tensorrt_llm::executor::samplingconfig::mtoppdecay (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mTopPDecayE", false]], "tensorrt_llm::executor::samplingconfig::mtoppmin (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig8mTopPMinE", false]], "tensorrt_llm::executor::samplingconfig::mtoppresetids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig13mTopPResetIdsE", false]], "tensorrt_llm::executor::samplingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", false]], "tensorrt_llm::executor::samplingconfig::samplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", false]], "tensorrt_llm::executor::samplingconfig::setbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::setbeamwidtharray (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", false]], "tensorrt_llm::executor::samplingconfig::setearlystopping (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setfrequencypenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setlengthpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setmintokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setnorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setnumreturnsequences (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setpresencepenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setrepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setseed (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settopk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::settopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppdecay (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppmin (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppresetids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", false]], "tensorrt_llm::executor::samplingconfig::updatenumreturnbeams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20updateNumReturnBeamsEv", false]], "tensorrt_llm::executor::schedulerconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfigE", false]], "tensorrt_llm::executor::schedulerconfig::getcapacityschedulerpolicy (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig26getCapacitySchedulerPolicyEv", false]], "tensorrt_llm::executor::schedulerconfig::getcontextchunkingpolicy (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig24getContextChunkingPolicyEv", false]], "tensorrt_llm::executor::schedulerconfig::getdynamicbatchconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig21getDynamicBatchConfigEv", false]], "tensorrt_llm::executor::schedulerconfig::mcapacityschedulerpolicy (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig24mCapacitySchedulerPolicyE", false]], "tensorrt_llm::executor::schedulerconfig::mcontextchunkingpolicy (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig22mContextChunkingPolicyE", false]], "tensorrt_llm::executor::schedulerconfig::mdynamicbatchconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig19mDynamicBatchConfigE", false]], "tensorrt_llm::executor::schedulerconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", false]], "tensorrt_llm::executor::schedulerconfig::schedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", false]], "tensorrt_llm::executor::serialization (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor13SerializationE", false]], "tensorrt_llm::executor::serialization::deserializeadditionalmodeloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeadditionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializebool (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecommstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecontextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedatatransceiverstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedebugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedecodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedisservingrequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedynamicbatchconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeeagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeexecutorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeguideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeguideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeinflightbatchingstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeiterationstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeiterationstatsvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializekvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializekvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializekvcachestats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializelookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeloraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializemodeltype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializemropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeorchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeoutputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeparallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializepeftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeprompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequestperfmetrics (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststage (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststatsperiteration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststatsperiterationvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializeresponse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializeresult (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializesamplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeschedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializesocketstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializespecdecfastlogitsinfo (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializespeculativedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializestaticbatchingstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializestring (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetimepoint (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetokenrangeretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::serialize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", false]], "tensorrt_llm::executor::serialization::serializedsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN18RequestPerfMetrics9TimePointE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", false]], "tensorrt_llm::executor::shape (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor5ShapeE", false]], "tensorrt_llm::executor::shape::base (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape4BaseE", false]], "tensorrt_llm::executor::shape::dimtype64 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape9DimType64E", false]], "tensorrt_llm::executor::shape::shape (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", false], [0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", false], [0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEv", false]], "tensorrt_llm::executor::sizetype32 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10SizeType32E", false]], "tensorrt_llm::executor::speculativedecodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfigE", false]], "tensorrt_llm::executor::speculativedecodingconfig::fastlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig10fastLogitsE", false]], "tensorrt_llm::executor::speculativedecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", false]], "tensorrt_llm::executor::speculativedecodingconfig::speculativedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfoE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::draftparticipantid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo18draftParticipantIdE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::draftrequestid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo14draftRequestIdE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::totensor (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo8toTensorEv", false]], "tensorrt_llm::executor::staticbatchingstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStatsE", false]], "tensorrt_llm::executor::staticbatchingstats::emptygenslots (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats13emptyGenSlotsE", false]], "tensorrt_llm::executor::staticbatchingstats::numcontextrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats18numContextRequestsE", false]], "tensorrt_llm::executor::staticbatchingstats::numctxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numCtxTokensE", false]], "tensorrt_llm::executor::staticbatchingstats::numgentokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numGenTokensE", false]], "tensorrt_llm::executor::staticbatchingstats::numscheduledrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats20numScheduledRequestsE", false]], "tensorrt_llm::executor::streamptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9StreamPtrE", false]], "tensorrt_llm::executor::tensor (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor6TensorE", false]], "tensorrt_llm::executor::tensor::copyto (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::copytocpu (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytogpu (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytomanaged (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytopinned (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytopooledpinned (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::cpu (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::cudastreamptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::detail::ofitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", false]], "tensorrt_llm::executor::tensor::detail::toitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", false]], "tensorrt_llm::executor::tensor::getdata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7getDataEv", false], [0, "_CPPv4NK12tensorrt_llm8executor6Tensor7getDataEv", false]], "tensorrt_llm::executor::tensor::getdatatype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor11getDataTypeEv", false]], "tensorrt_llm::executor::tensor::getmemorytype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor13getMemoryTypeEv", false]], "tensorrt_llm::executor::tensor::getruntimetype (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", false]], "tensorrt_llm::executor::tensor::getshape (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor8getShapeEv", false]], "tensorrt_llm::executor::tensor::getsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor7getSizeEv", false]], "tensorrt_llm::executor::tensor::getsizeinbytes (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor14getSizeInBytesEv", false]], "tensorrt_llm::executor::tensor::gpu (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", false]], "tensorrt_llm::executor::tensor::impl (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor4ImplE", false]], "tensorrt_llm::executor::tensor::managed (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::mtensor (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7mTensorE", false]], "tensorrt_llm::executor::tensor::of (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", false], [0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", false]], "tensorrt_llm::executor::tensor::operator bool (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensorcvbEv", false]], "tensorrt_llm::executor::tensor::operator!= (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", false]], "tensorrt_llm::executor::tensor::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", false]], "tensorrt_llm::executor::tensor::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", false]], "tensorrt_llm::executor::tensor::pinned (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::pooledpinned (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::setfrom (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::setzero (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::tensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorEv", false]], "tensorrt_llm::executor::tensor::~tensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6TensorD0Ev", false]], "tensorrt_llm::executor::tensorptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9TensorPtrE", false]], "tensorrt_llm::executor::tokenidtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11TokenIdTypeE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIbEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIbE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIfEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIfE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsI4halfEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsI4halfE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int32_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int32_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int64_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int64_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt6int8_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt6int8_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIP1TE5valueE", false]], "tensorrt_llm::executor::veclogprobs (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11VecLogProbsE", false]], "tensorrt_llm::executor::vectokenextraids (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16VecTokenExtraIdsE", false]], "tensorrt_llm::executor::vectokens (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9VecTokensE", false]], "tensorrt_llm::executor::version (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7versionEv", false]], "tensorrt_llm::layers (c++ type)": [[1, "_CPPv4N12tensorrt_llm6layersE", false]], "tensorrt_llm::mpi (c++ type)": [[0, "_CPPv4N12tensorrt_llm3mpiE", false]], "tensorrt_llm::runtime (c++ type)": [[0, "_CPPv4N12tensorrt_llm7runtimeE", false], [1, "_CPPv4N12tensorrt_llm7runtimeE", false]], "tensorrt_llm::runtime::allreducebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffersE", false]], "tensorrt_llm::runtime::allreducebuffers::allreducebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", false]], "tensorrt_llm::runtime::allreducebuffers::mallreducecommptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers18mAllReduceCommPtrsE", false]], "tensorrt_llm::runtime::allreducebuffers::mipcmemoryhandles (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers17mIpcMemoryHandlesE", false]], "tensorrt_llm::runtime::allreducebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::buffercast (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", false]], "tensorrt_llm::runtime::buffercastornull (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", false]], "tensorrt_llm::runtime::bufferdatatype (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::bufferdatatype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", false]], "tensorrt_llm::runtime::bufferdatatype::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType11getDataTypeEv", false]], "tensorrt_llm::runtime::bufferdatatype::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType7getSizeEv", false]], "tensorrt_llm::runtime::bufferdatatype::getsizeinbits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType13getSizeInBitsEv", false]], "tensorrt_llm::runtime::bufferdatatype::ispointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType9isPointerEv", false]], "tensorrt_llm::runtime::bufferdatatype::isunsigned (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType10isUnsignedEv", false]], "tensorrt_llm::runtime::bufferdatatype::ktrtpointertype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType15kTrtPointerTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mDataTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::mpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType8mPointerE", false]], "tensorrt_llm::runtime::bufferdatatype::munsigned (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mUnsignedE", false]], "tensorrt_llm::runtime::bufferdatatype::operator nvinfer1::datatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataTypecvN8nvinfer18DataTypeEEv", false]], "tensorrt_llm::runtime::buffermanager (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerE", false]], "tensorrt_llm::runtime::buffermanager::allocate (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::buffermanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", false]], "tensorrt_llm::runtime::buffermanager::copy (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", false]], "tensorrt_llm::runtime::buffermanager::copyfrom (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", false]], "tensorrt_llm::runtime::buffermanager::cpu (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::cudamempoolptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager14CudaMemPoolPtrE", false]], "tensorrt_llm::runtime::buffermanager::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13CudaStreamPtrE", false]], "tensorrt_llm::runtime::buffermanager::emptybuffer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::emptytensor (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::getstream (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager9getStreamEv", false]], "tensorrt_llm::runtime::buffermanager::gpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::gpusync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::ibufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10IBufferPtrE", false]], "tensorrt_llm::runtime::buffermanager::ipcnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::itensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10ITensorPtrE", false]], "tensorrt_llm::runtime::buffermanager::kbyte_type (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10kBYTE_TYPEE", false]], "tensorrt_llm::runtime::buffermanager::managed (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::memorypoolfree (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolFreeEv", false]], "tensorrt_llm::runtime::buffermanager::memorypoolreserved (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager18memoryPoolReservedEv", false]], "tensorrt_llm::runtime::buffermanager::memorypooltrimto (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", false]], "tensorrt_llm::runtime::buffermanager::memorypoolused (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolUsedEv", false]], "tensorrt_llm::runtime::buffermanager::mpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager5mPoolE", false]], "tensorrt_llm::runtime::buffermanager::mstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7mStreamE", false]], "tensorrt_llm::runtime::buffermanager::mtrimpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager9mTrimPoolE", false]], "tensorrt_llm::runtime::buffermanager::pinned (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::pinnedpool (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::setmem (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", false]], "tensorrt_llm::runtime::buffermanager::setzero (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", false]], "tensorrt_llm::runtime::buffermanager::~buffermanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerD0Ev", false]], "tensorrt_llm::runtime::bufferrange (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", false]], "tensorrt_llm::runtime::bufferrange::base (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11BufferRange4BaseE", false]], "tensorrt_llm::runtime::bufferrange::bufferrange (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", false], [1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", false]], "tensorrt_llm::runtime::canaccesspeer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", false]], "tensorrt_llm::runtime::constpointercast (c++ function)": [[1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", false]], "tensorrt_llm::runtime::cudaevent (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEventE", false]], "tensorrt_llm::runtime::cudaevent::cudaevent (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", false], [1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", false]], "tensorrt_llm::runtime::cudaevent::deleter (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7DeleterE", false]], "tensorrt_llm::runtime::cudaevent::deleter::deleter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", false], [1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEv", false]], "tensorrt_llm::runtime::cudaevent::deleter::mownsevent (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter10mOwnsEventE", false]], "tensorrt_llm::runtime::cudaevent::deleter::operator() (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", false]], "tensorrt_llm::runtime::cudaevent::element_type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent12element_typeE", false]], "tensorrt_llm::runtime::cudaevent::eventptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent8EventPtrE", false]], "tensorrt_llm::runtime::cudaevent::get (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent3getEv", false]], "tensorrt_llm::runtime::cudaevent::mevent (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent6mEventE", false]], "tensorrt_llm::runtime::cudaevent::pointer (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7pointerE", false]], "tensorrt_llm::runtime::cudaevent::synchronize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent11synchronizeEv", false]], "tensorrt_llm::runtime::cudastream (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStreamE", false]], "tensorrt_llm::runtime::cudastream::cudastream (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", false]], "tensorrt_llm::runtime::cudastream::deleter (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7DeleterE", false]], "tensorrt_llm::runtime::cudastream::deleter::deleter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEv", false]], "tensorrt_llm::runtime::cudastream::deleter::mownsstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter11mOwnsStreamE", false]], "tensorrt_llm::runtime::cudastream::deleter::operator() (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", false]], "tensorrt_llm::runtime::cudastream::get (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream3getEv", false]], "tensorrt_llm::runtime::cudastream::getdevice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream9getDeviceEv", false]], "tensorrt_llm::runtime::cudastream::mdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mDeviceE", false]], "tensorrt_llm::runtime::cudastream::mstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mStreamE", false]], "tensorrt_llm::runtime::cudastream::record (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", false]], "tensorrt_llm::runtime::cudastream::streamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream9StreamPtrE", false]], "tensorrt_llm::runtime::cudastream::synchronize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream11synchronizeEv", false]], "tensorrt_llm::runtime::cudastream::wait (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4typeE", false]], "tensorrt_llm::runtime::decoder (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoderE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffersE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::beamsearchbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::mcumlogprobstmp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers15mCumLogProbsTmpE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::mnumsms (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7mNumSMsE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::moutputbeamhypotheses (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers21mOutputBeamHypothesesE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decoder::decoderstate (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderStateE", false]], "tensorrt_llm::runtime::decoder::decoderstate::allocatespeculativedecodingbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::decoderstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::decodinginputptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16DecodingInputPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::decodingoutputptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState17DecodingOutputPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", false]], "tensorrt_llm::runtime::decoder::decoderstate::getacceptedlengthscumsum (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24getAcceptedLengthsCumSumEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getacceptedpackedpaths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getAcceptedPackedPathsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getallnewtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getAllNewTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getbeamsearchbuffers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getBeamSearchBuffersEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getcumlogprobs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishedsteps (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishedStepsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishedsum (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getFinishedSumEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishreasons (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishReasonsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getgatheredids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getjointdecodinginput (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState21getJointDecodingInputEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getjointdecodingoutput (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getJointDecodingOutputEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getlogprobs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxdecodingdecodertokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState27getMaxDecodingDecoderTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxdecodingenginetokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getMaxDecodingEngineTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxsequencelength (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getMaxSequenceLengthEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnextdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getNextDraftTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnextdrafttokenslengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getNextDraftTokensLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnumdecodingenginetokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getparentids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState12getParentIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getprevdrafttokenslengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getPrevDraftTokensLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getsequencelengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getspeculativedecodingmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getSpeculativeDecodingModeEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13LlmRequestPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mbeamsearchbuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mBeamSearchBuffersE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mfinishedsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState14mFinishedStepsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mjointdecodinginput (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState19mJointDecodingInputE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mjointdecodingoutput (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState20mJointDecodingOutputE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxbeamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBeamWidthE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxdecodingdecodertokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState25mMaxDecodingDecoderTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxdecodingenginetokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mMaxDecodingEngineTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxsequencelength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mMaxSequenceLengthE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mnumdecodingenginetokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mNumDecodingEngineTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mspeculativedecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mSpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::decoder::decoderstate::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13RequestVectorE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setnumdecodingenginetokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decoder::decoderstate::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupeagle (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupexplicitdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setuplookahead (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupspeculativedecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", false]], "tensorrt_llm::runtime::decoder_batch::input (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5InputE", false]], "tensorrt_llm::runtime::decoder_batch::input::batchslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input10batchSlotsE", false]], "tensorrt_llm::runtime::decoder_batch::input::batchslotsrequestorder (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input22batchSlotsRequestOrderE", false]], "tensorrt_llm::runtime::decoder_batch::input::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input16cacheIndirectionE", false]], "tensorrt_llm::runtime::decoder_batch::input::eagleinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input11eagleInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::eaglelastinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15eagleLastInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::explicitdrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input25explicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::explicitdrafttokenslastinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input29explicitDraftTokensLastInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::generationsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15generationStepsE", false]], "tensorrt_llm::runtime::decoder_batch::input::input (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", false], [1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", false]], "tensorrt_llm::runtime::decoder_batch::input::logits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input6logitsE", false]], "tensorrt_llm::runtime::decoder_batch::input::maxdecodersteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15maxDecoderStepsE", false]], "tensorrt_llm::runtime::decoder_batch::input::predicteddraftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input20predictedDraftLogitsE", false]], "tensorrt_llm::runtime::decoder_batch::input::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input14TensorConstPtrE", false]], "tensorrt_llm::runtime::decoder_batch::input::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch::output (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6OutputE", false]], "tensorrt_llm::runtime::decoder_batch::output::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output16cacheIndirectionE", false]], "tensorrt_llm::runtime::decoder_batch::output::output (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output6OutputEv", false]], "tensorrt_llm::runtime::decoder_batch::output::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7RequestE", false]], "tensorrt_llm::runtime::decoder_batch::request::badwordslist (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12badWordsListE", false]], "tensorrt_llm::runtime::decoder_batch::request::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9BufferPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request::draftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftLogitsE", false]], "tensorrt_llm::runtime::decoder_batch::request::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftTokensE", false]], "tensorrt_llm::runtime::decoder_batch::request::dtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5dtypeE", false]], "tensorrt_llm::runtime::decoder_batch::request::eagleconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11eagleConfigE", false]], "tensorrt_llm::runtime::decoder_batch::request::embeddingbias (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13embeddingBiasE", false]], "tensorrt_llm::runtime::decoder_batch::request::endid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5endIdE", false]], "tensorrt_llm::runtime::decoder_batch::request::generatedtokensperenginestep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request28generatedTokensPerEngineStepE", false]], "tensorrt_llm::runtime::decoder_batch::request::ids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request3idsE", false]], "tensorrt_llm::runtime::decoder_batch::request::inputlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request8inputLenE", false]], "tensorrt_llm::runtime::decoder_batch::request::lookaheadruntimeconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request22lookaheadRuntimeConfigE", false]], "tensorrt_llm::runtime::decoder_batch::request::maxnewtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12maxNewTokensE", false]], "tensorrt_llm::runtime::decoder_batch::request::medusapaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11medusaPathsE", false]], "tensorrt_llm::runtime::decoder_batch::request::medusatreeids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13medusaTreeIdsE", false]], "tensorrt_llm::runtime::decoder_batch::request::request (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::decoder_batch::request::stopwordslist (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13stopWordsListE", false]], "tensorrt_llm::runtime::decoder_batch::request::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request14TensorConstPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9TensorPtrE", false]], "tensorrt_llm::runtime::decodinginput (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInputE", false]], "tensorrt_llm::runtime::decodinginput::badwordslens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsLensE", false]], "tensorrt_llm::runtime::decodinginput::badwordslists (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13badWordsListsE", false]], "tensorrt_llm::runtime::decodinginput::badwordsptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsPtrsE", false]], "tensorrt_llm::runtime::decodinginput::batchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9batchSizeE", false]], "tensorrt_llm::runtime::decodinginput::batchslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10batchSlotsE", false]], "tensorrt_llm::runtime::decodinginput::beamwidths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10beamWidthsE", false]], "tensorrt_llm::runtime::decodinginput::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput16cacheIndirectionE", false]], "tensorrt_llm::runtime::decodinginput::decodinginput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11eagleInputsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs12acceptedLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedpathids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15acceptedPathIdsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14acceptedTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::eagleinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13lastDraftLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14lastDraftPathsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15lastDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13nextDraftLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14nextDraftPathsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::seqslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs8seqSlotsE", false]], "tensorrt_llm::runtime::decodinginput::embeddingbias (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13embeddingBiasE", false]], "tensorrt_llm::runtime::decodinginput::endids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6endIdsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25explicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::bestpathindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::bestpathlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16lastDraftIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15lastDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21lastGenerationLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastpositionidsbase (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs19lastPositionIdsBaseE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::masks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs5masksE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::maxgenlengthdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs18maxGenLengthDeviceE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16nextDraftIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdraftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextDraftProbsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextflattokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextFlatTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21nextGenerationLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::packedpositionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs17packedPositionIdsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::seqslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs8seqSlotsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25externalDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::constantthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs17constantThresholdE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::draftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11draftLogitsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::draftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs10draftProbsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::drafttokenids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs13draftTokenIdsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::numdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14numDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::numdrafttokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18numDraftTokensHostE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::step (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs4stepE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::targetprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11targetProbsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::usedraftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14useDraftLogitsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::usedraftlogitshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18useDraftLogitsHostE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::userandomacceptancethreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs28useRandomAcceptanceThresholdE", false]], "tensorrt_llm::runtime::decodinginput::finishreasons (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13finishReasonsE", false]], "tensorrt_llm::runtime::decodinginput::generationsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15generationStepsE", false]], "tensorrt_llm::runtime::decodinginput::lengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput7lengthsE", false]], "tensorrt_llm::runtime::decodinginput::logits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6logitsE", false]], "tensorrt_llm::runtime::decodinginput::logitsvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9logitsVecE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15lookaheadInputsE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputsE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs::tokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputs13tokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::maxattentionwindow (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput18maxAttentionWindowE", false]], "tensorrt_llm::runtime::decodinginput::maxbadwordslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14maxBadWordsLenE", false]], "tensorrt_llm::runtime::decodinginput::maxlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9maxLengthE", false]], "tensorrt_llm::runtime::decodinginput::maxstopwordslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15maxStopWordsLenE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12medusaInputsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusacurtokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs22medusaCurTokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusalogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs12medusaLogitsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusapaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs11medusaPathsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusatargettokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs25medusaTargetTokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusatreeids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs13medusaTreeIdsE", false]], "tensorrt_llm::runtime::decodinginput::norepeatngramsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput17noRepeatNgramSizeE", false]], "tensorrt_llm::runtime::decodinginput::sequencelimitlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput19sequenceLimitLengthE", false]], "tensorrt_llm::runtime::decodinginput::sinktokenlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15sinkTokenLengthE", false]], "tensorrt_llm::runtime::decodinginput::step (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput4stepE", false]], "tensorrt_llm::runtime::decodinginput::stopwordslens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsLensE", false]], "tensorrt_llm::runtime::decodinginput::stopwordslists (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14stopWordsListsE", false]], "tensorrt_llm::runtime::decodinginput::stopwordsptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsPtrsE", false]], "tensorrt_llm::runtime::decodinginput::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14TensorConstPtrE", false]], "tensorrt_llm::runtime::decodinginput::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9TensorPtrE", false]], "tensorrt_llm::runtime::decodingoutput (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutputE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypothesesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14beamHypothesesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::batchdones (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses10batchDonesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::cumlogprobscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses14cumLogProbsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::empty (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::init (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::logprobscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11logProbsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::minnormedscorescba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18minNormedScoresCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::normedscorescba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses15normedScoresCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::numbeamscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11numBeamsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::outputidscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses12outputIdsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::release (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7releaseEv", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::sequencelengthscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18sequenceLengthsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::slice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decodingoutput::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16cacheIndirectionE", false]], "tensorrt_llm::runtime::decodingoutput::cumlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11cumLogProbsE", false]], "tensorrt_llm::runtime::decodingoutput::decodingoutput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::decodingoutput::eaglebuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12eagleBuffersE", false]], "tensorrt_llm::runtime::decodingoutput::explicitdrafttokensbuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26explicitDraftTokensBuffersE", false]], "tensorrt_llm::runtime::decodingoutput::finishedsum (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11finishedSumE", false]], "tensorrt_llm::runtime::decodingoutput::finishreasons (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13finishReasonsE", false]], "tensorrt_llm::runtime::decodingoutput::gatheredids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11gatheredIdsE", false]], "tensorrt_llm::runtime::decodingoutput::ids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput3idsE", false]], "tensorrt_llm::runtime::decodingoutput::knegativeinfinity (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput17kNegativeInfinityE", false]], "tensorrt_llm::runtime::decodingoutput::lengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput7lengthsE", false]], "tensorrt_llm::runtime::decodingoutput::logprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput8logProbsE", false]], "tensorrt_llm::runtime::decodingoutput::logprobstiled (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13logProbsTiledE", false]], "tensorrt_llm::runtime::decodingoutput::lookaheadoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16lookaheadOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::newtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9newTokensE", false]], "tensorrt_llm::runtime::decodingoutput::newtokenssteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14newTokensStepsE", false]], "tensorrt_llm::runtime::decodingoutput::newtokensvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12newTokensVecE", false]], "tensorrt_llm::runtime::decodingoutput::parentids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9parentIdsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26speculativeDecodingOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::acceptedlengthscumsum (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs21acceptedLengthsCumSumE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::acceptedtokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs17acceptedTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::nextdrafttokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18nextDraftTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::pathsoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs12pathsOffsetsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::prevdrafttokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18prevDraftTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9TensorPtrE", false]], "tensorrt_llm::runtime::deviceallocationnvls (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", false]], "tensorrt_llm::runtime::deviceallocationnvls::_capacity (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls9_capacityE", false]], "tensorrt_llm::runtime::deviceallocationnvls::_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls7_handleE", false]], "tensorrt_llm::runtime::deviceallocationnvls::deviceallocationnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls20DeviceAllocationNvlsEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::free (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls4freeEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getcapacity (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls11getCapacityEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getipcunicastpointers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls21getIpcUnicastPointersEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getmulticastpointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls19getMulticastPointerEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getunicastpointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls17getUnicastPointerEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::reset (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", false]], "tensorrt_llm::runtime::deviceallocationnvls::~deviceallocationnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvlsD0Ev", false]], "tensorrt_llm::runtime::eaglebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffersE", false]], "tensorrt_llm::runtime::eaglebuffers::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9BufferPtrE", false]], "tensorrt_llm::runtime::eaglebuffers::chunkedcontextnexttokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers28chunkedContextNextTokensHostE", false]], "tensorrt_llm::runtime::eaglebuffers::cumsumgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers23cumSumGenerationLengthsE", false]], "tensorrt_llm::runtime::eaglebuffers::eaglebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::engineinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12engineInputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13engineOutputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs12acceptedLensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13acceptedPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14acceptedTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13nextDraftLensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14nextDraftPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::greedysamplinghost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18greedySamplingHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6InputsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersdrafttokenids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22allLayersDraftTokenIdsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersdrafttokenidspredecessor (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33allLayersDraftTokenIdsPredecessorE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersscores (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15allLayersScoresE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::currentexpandindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20currentExpandIndicesE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs9draftLensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10draftPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftpathshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14draftPathsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs11draftTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::dynamictreemaxtopkhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22dynamicTreeMaxTopKHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxcontextlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetCtxContextLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxpastkeyvaluelengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetCtxPastKeyValueLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxrequesttypeshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetCtxRequestTypesHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgencontextlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetGenContextLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgenpastkeyvaluelengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetGenPastKeyValueLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgenrequesttypeshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetGenRequestTypesHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::inputgentokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18inputGenTokensHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::posterioralpha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14posteriorAlphaE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::posteriorthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18posteriorThresholdE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::prevscores (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10prevScoresE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::randomdatasample (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs16randomDataSampleE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::randomdatavalidation (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20randomDataValidationE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodinggenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29specDecodingGenerationLengthsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodinggenerationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33specDecodingGenerationLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodingpackedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs23specDecodingPackedMasksE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodingpositionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27specDecodingPositionOffsetsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::temperatures (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs12temperaturesE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::usedynamictreehost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18useDynamicTreeHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15useSpecDecodingE", false]], "tensorrt_llm::runtime::eaglebuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::itensor (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7ITensorE", false]], "tensorrt_llm::runtime::eaglebuffers::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13LlmRequestPtrE", false]], "tensorrt_llm::runtime::eaglebuffers::maxgenerationlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers19maxGenerationLengthE", false]], "tensorrt_llm::runtime::eaglebuffers::mdefaultposteriorthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26mDefaultPosteriorThresholdE", false]], "tensorrt_llm::runtime::eaglebuffers::mdogreedysampling (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers17mDoGreedySamplingE", false]], "tensorrt_llm::runtime::eaglebuffers::posterioralphahost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18posteriorAlphaHostE", false]], "tensorrt_llm::runtime::eaglebuffers::posteriorthresholdhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers22posteriorThresholdHostE", false]], "tensorrt_llm::runtime::eaglebuffers::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13RequestVectorE", false]], "tensorrt_llm::runtime::eaglebuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::scanreducetempstorage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers21scanReduceTempStorageE", false]], "tensorrt_llm::runtime::eaglebuffers::scanreducetempstoragebytes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26scanReduceTempStorageBytesE", false]], "tensorrt_llm::runtime::eaglebuffers::setfrominputs (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers10SizeType32E", false]], "tensorrt_llm::runtime::eaglebuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorMapE", false]], "tensorrt_llm::runtime::eaglebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::eaglemodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModuleE", false]], "tensorrt_llm::runtime::eaglemodule::eaglemodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleEv", false]], "tensorrt_llm::runtime::eaglemodule::getdefaulteaglechoices (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule22getDefaultEagleChoicesEv", false]], "tensorrt_llm::runtime::eaglemodule::getmaxnonleafnodesperlayer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule26getMaxNonLeafNodesPerLayerEv", false]], "tensorrt_llm::runtime::eaglemodule::getnumtransformerlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule23getNumTransformerLayersEv", false]], "tensorrt_llm::runtime::eaglemodule::mdefaulteaglechoices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule20mDefaultEagleChoicesE", false]], "tensorrt_llm::runtime::eaglemodule::mmaxnonleafnodesperlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule24mMaxNonLeafNodesPerLayerE", false]], "tensorrt_llm::runtime::eaglemodule::mnumtransformerslayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule21mNumTransformersLayerE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffersE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9BufferPtrE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::cumsumgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers23cumSumGenerationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12engineInputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs::positionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs15positionOffsetsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs::requesttypesdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs18requestTypesDeviceE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13engineOutputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::bestpathindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::bestpathlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::masks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs5masksE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::maxgentoken (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs11maxGenTokenE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs16nextDraftIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdraftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextDraftProbsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextflattokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextFlatTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs21nextGenerationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextpositionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs19nextPositionOffsetsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::packedpositionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs17packedPositionIdsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::totalgentoken (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs13totalGenTokenE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::explicitdrafttokensbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6InputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::draftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12draftIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::draftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs10draftProbsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11draftTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::generationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs17generationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::generationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs21generationLengthsHostE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::maxgenlengthhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16maxGenLengthHostE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::packedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11packedMasksE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::positionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11positionIdsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::positionidsbase (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15positionIdsBaseE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::randomdatasample (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16randomDataSampleE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::randomdatavalidation (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs20randomDataValidationE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::temperatures (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12temperaturesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15useSpecDecodingE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::itensor (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7ITensorE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::scantempstorage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers15scanTempStorageE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::scantempstoragebytes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers20scanTempStorageBytesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::setfrominputs (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", false], [1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers10SizeType32E", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorMapE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::genericprompttuningparams (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", false]], "tensorrt_llm::runtime::genericprompttuningparams::embeddingtable (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams14embeddingTableE", false]], "tensorrt_llm::runtime::genericprompttuningparams::genericprompttuningparams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::genericprompttuningparams::prompttuningenabled (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams19promptTuningEnabledE", false]], "tensorrt_llm::runtime::genericprompttuningparams::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams10SizeType32E", false]], "tensorrt_llm::runtime::genericprompttuningparams::tasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams5tasksE", false]], "tensorrt_llm::runtime::genericprompttuningparams::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9TensorPtrE", false]], "tensorrt_llm::runtime::genericprompttuningparams::vocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9vocabSizeE", false]], "tensorrt_llm::runtime::getdefaultbatchslots (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", false]], "tensorrt_llm::runtime::gptdecoder (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", false]], "tensorrt_llm::runtime::gptdecoder::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13CudaStreamPtrE", false]], "tensorrt_llm::runtime::gptdecoder::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", false]], "tensorrt_llm::runtime::gptdecoder::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::gptdecoder::forwardsync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::gptdecoder::getsamplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder17getSamplingConfigEv", false]], "tensorrt_llm::runtime::gptdecoder::gptdecoder (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::gptdecoder::mdecodinglayerworkspace (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder23mDecodingLayerWorkspaceE", false]], "tensorrt_llm::runtime::gptdecoder::mdecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mDecodingModeE", false]], "tensorrt_llm::runtime::gptdecoder::mdynamicdecodelayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder19mDynamicDecodeLayerE", false]], "tensorrt_llm::runtime::gptdecoder::mmanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder8mManagerE", false]], "tensorrt_llm::runtime::gptdecoder::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::gptdecoder::msamplingconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder15mSamplingConfigE", false]], "tensorrt_llm::runtime::gptdecoder::mvocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10mVocabSizeE", false]], "tensorrt_llm::runtime::gptdecoder::mvocabsizepadded (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16mVocabSizePaddedE", false]], "tensorrt_llm::runtime::gptdecoder::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", false]], "tensorrt_llm::runtime::gptdecoder::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder9TensorPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatchedE", false]], "tensorrt_llm::runtime::gptdecoderbatched::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13CudaStreamPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", false]], "tensorrt_llm::runtime::gptdecoderbatched::finalize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", false]], "tensorrt_llm::runtime::gptdecoderbatched::forward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::forwarddispatch (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::getbuffermanager (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getBufferManagerEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getdecoderstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getdecoderstream (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getDecoderStreamEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getunderlyingdecoder (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched20getUnderlyingDecoderEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::gptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::gptdecoderbatched::gptdecoderptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13GptDecoderPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13LlmRequestPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mbuffermanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mBufferManagerE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoder (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched8mDecoderE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoderstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13mDecoderStateE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoderstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mDecoderStreamE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mruntimestream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mRuntimeStreamE", false]], "tensorrt_llm::runtime::gptdecoderbatched::prepareforward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13RequestVectorE", false]], "tensorrt_llm::runtime::gptdecoderbatched::seteagleinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::setexplicitdrafttokensinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::gptdecoderbatched::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14SharedConstPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched9TensorPtrE", false]], "tensorrt_llm::runtime::gptjsonconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfigE", false]], "tensorrt_llm::runtime::gptjsonconfig::enginefilename (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", false], [1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", false]], "tensorrt_llm::runtime::gptjsonconfig::getcontextparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig21getContextParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getgpuspernode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getGpusPerNodeEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getmodelconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getModelConfigEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getmodelconfigmutable (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig21getModelConfigMutableEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getname (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig7getNameEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getpipelineparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig22getPipelineParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getprecision (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getPrecisionEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getruntimedefaults (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig18getRuntimeDefaultsEv", false]], "tensorrt_llm::runtime::gptjsonconfig::gettensorparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig20getTensorParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getversion (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig10getVersionEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getworldsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getWorldSizeEv", false]], "tensorrt_llm::runtime::gptjsonconfig::gptjsonconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", false]], "tensorrt_llm::runtime::gptjsonconfig::mcontextparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig19mContextParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mGpusPerNodeE", false]], "tensorrt_llm::runtime::gptjsonconfig::mmodelconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mModelConfigE", false]], "tensorrt_llm::runtime::gptjsonconfig::mname (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5mNameE", false]], "tensorrt_llm::runtime::gptjsonconfig::mpipelineparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig20mPipelineParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mprecision (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig10mPrecisionE", false]], "tensorrt_llm::runtime::gptjsonconfig::mruntimedefaults (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig16mRuntimeDefaultsE", false]], "tensorrt_llm::runtime::gptjsonconfig::mtensorparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig18mTensorParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mversion (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig8mVersionE", false]], "tensorrt_llm::runtime::gptjsonconfig::parse (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", false], [1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", false], [1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", false]], "tensorrt_llm::runtime::ibuffer (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferE", false]], "tensorrt_llm::runtime::ibuffer::data (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataEv", false]], "tensorrt_llm::runtime::ibuffer::datatype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer8DataTypeE", false]], "tensorrt_llm::runtime::ibuffer::getcapacity (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getCapacityEv", false]], "tensorrt_llm::runtime::ibuffer::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getDataTypeEv", false]], "tensorrt_llm::runtime::ibuffer::getdatatypename (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer15getDataTypeNameEv", false]], "tensorrt_llm::runtime::ibuffer::getmemorytype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer13getMemoryTypeEv", false]], "tensorrt_llm::runtime::ibuffer::getmemorytypename (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer17getMemoryTypeNameEv", false]], "tensorrt_llm::runtime::ibuffer::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7getSizeEv", false]], "tensorrt_llm::runtime::ibuffer::getsizeinbytes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer14getSizeInBytesEv", false]], "tensorrt_llm::runtime::ibuffer::ibuffer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferERK7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferEv", false]], "tensorrt_llm::runtime::ibuffer::memorytype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", false]], "tensorrt_llm::runtime::ibuffer::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferaSERK7IBuffer", false]], "tensorrt_llm::runtime::ibuffer::release (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7releaseEv", false]], "tensorrt_llm::runtime::ibuffer::resize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14SharedConstPtrE", false]], "tensorrt_llm::runtime::ibuffer::sharedptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9SharedPtrE", false]], "tensorrt_llm::runtime::ibuffer::slice (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::tobytes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::uniqueconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14UniqueConstPtrE", false]], "tensorrt_llm::runtime::ibuffer::uniqueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9UniquePtrE", false]], "tensorrt_llm::runtime::ibuffer::view (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::wrap (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::~ibuffer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferD0Ev", false]], "tensorrt_llm::runtime::igptdecoder (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderE", false]], "tensorrt_llm::runtime::igptdecoder::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::igptdecoder::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", false]], "tensorrt_llm::runtime::igptdecoder::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::igptdecoder::forwardsync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::igptdecoder::getsamplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder17getSamplingConfigEv", false]], "tensorrt_llm::runtime::igptdecoder::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", false]], "tensorrt_llm::runtime::igptdecoder::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder14TensorConstPtrE", false]], "tensorrt_llm::runtime::igptdecoder::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder9TensorPtrE", false]], "tensorrt_llm::runtime::igptdecoder::~igptdecoder (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderD0Ev", false]], "tensorrt_llm::runtime::igptdecoderbatched (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedE", false]], "tensorrt_llm::runtime::igptdecoderbatched::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13CudaStreamPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", false]], "tensorrt_llm::runtime::igptdecoderbatched::finalize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", false]], "tensorrt_llm::runtime::igptdecoderbatched::forward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::igptdecoderbatched::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::igptdecoderbatched::igptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched18IGptDecoderBatchedEv", false]], "tensorrt_llm::runtime::igptdecoderbatched::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13LlmRequestPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13RequestVectorE", false]], "tensorrt_llm::runtime::igptdecoderbatched::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::igptdecoderbatched::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched9TensorPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::~igptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedD0Ev", false]], "tensorrt_llm::runtime::ipcmemory (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryE", false]], "tensorrt_llm::runtime::ipcmemory::allocateipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", false]], "tensorrt_llm::runtime::ipcmemory::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9BufferPtrE", false]], "tensorrt_llm::runtime::ipcmemory::destroyipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory16destroyIpcMemoryEv", false]], "tensorrt_llm::runtime::ipcmemory::flags_size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory10FLAGS_SIZEE", false]], "tensorrt_llm::runtime::ipcmemory::getcommptrs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9IpcMemory11getCommPtrsEv", false]], "tensorrt_llm::runtime::ipcmemory::ipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERK9IpcMemory", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERR9IpcMemory", false]], "tensorrt_llm::runtime::ipcmemory::mbuffer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mBufferE", false]], "tensorrt_llm::runtime::ipcmemory::mcommptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9mCommPtrsE", false]], "tensorrt_llm::runtime::ipcmemory::mopenipc (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory8mOpenIpcE", false]], "tensorrt_llm::runtime::ipcmemory::mtprank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mTpRankE", false]], "tensorrt_llm::runtime::ipcmemory::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERK9IpcMemory", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERR9IpcMemory", false]], "tensorrt_llm::runtime::ipcmemory::~ipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryD0Ev", false]], "tensorrt_llm::runtime::ipcnvlsallocate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", false]], "tensorrt_llm::runtime::ipcnvlsfree (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", false]], "tensorrt_llm::runtime::ipcnvlshandle (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_handles (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle14ipc_uc_handlesE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_ptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle11ipc_uc_ptrsE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_vas (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle10ipc_uc_vasE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9mc_handleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_ptr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6mc_ptrE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_va (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5mc_vaE", false]], "tensorrt_llm::runtime::ipcnvlshandle::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle4sizeE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9uc_handleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_ptr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6uc_ptrE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_va (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5uc_vaE", false]], "tensorrt_llm::runtime::ipcnvlssupported (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime16ipcNvlsSupportedEv", false]], "tensorrt_llm::runtime::itensor (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensorE", false]], "tensorrt_llm::runtime::itensor::at (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", false]], "tensorrt_llm::runtime::itensor::castsize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", false]], "tensorrt_llm::runtime::itensor::dimtype64 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9DimType64E", false]], "tensorrt_llm::runtime::itensor::flattenn (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", false]], "tensorrt_llm::runtime::itensor::getdimension (c++ function)": [[1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", false]], "tensorrt_llm::runtime::itensor::getshape (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7ITensor8getShapeEv", false]], "tensorrt_llm::runtime::itensor::itensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorERK7ITensor", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorEv", false]], "tensorrt_llm::runtime::itensor::makeshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", false]], "tensorrt_llm::runtime::itensor::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensoraSERK7ITensor", false]], "tensorrt_llm::runtime::itensor::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", false]], "tensorrt_llm::runtime::itensor::resize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", false]], "tensorrt_llm::runtime::itensor::shape (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor5ShapeE", false]], "tensorrt_llm::runtime::itensor::shapeequals (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", false], [1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", false], [1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", false]], "tensorrt_llm::runtime::itensor::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor14SharedConstPtrE", false]], "tensorrt_llm::runtime::itensor::sharedptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9SharedPtrE", false]], "tensorrt_llm::runtime::itensor::slice (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", false]], "tensorrt_llm::runtime::itensor::squeeze (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", false]], "tensorrt_llm::runtime::itensor::strides (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", false]], "tensorrt_llm::runtime::itensor::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9TensorMapE", false]], "tensorrt_llm::runtime::itensor::tostring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", false]], "tensorrt_llm::runtime::itensor::uniqueconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor14UniqueConstPtrE", false]], "tensorrt_llm::runtime::itensor::uniqueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9UniquePtrE", false]], "tensorrt_llm::runtime::itensor::unsqueeze (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", false]], "tensorrt_llm::runtime::itensor::view (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", false]], "tensorrt_llm::runtime::itensor::volume (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", false]], "tensorrt_llm::runtime::itensor::volumenonnegative (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", false]], "tensorrt_llm::runtime::itensor::wrap (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", false]], "tensorrt_llm::runtime::itensor::~itensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensorD0Ev", false]], "tensorrt_llm::runtime::lamportinitializeall (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffersE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::generationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers17generationLengthsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::lookaheaddecodingbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::packedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11packedMasksE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::positionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11positionIdsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::positionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers15positionOffsetsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::lookaheadmodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModuleE", false]], "tensorrt_llm::runtime::lookaheadmodule::getexecutionconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime15LookaheadModule18getExecutionConfigEv", false]], "tensorrt_llm::runtime::lookaheadmodule::lookaheadmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleEv", false]], "tensorrt_llm::runtime::lookaheadmodule::mexecutionconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule16mExecutionConfigE", false]], "tensorrt_llm::runtime::lookaheadmodule::setexecutionconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffersE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::batchslotshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18batchSlotsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::cumsumlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers12cumSumLengthE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::disablelookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers24disableLookaheadDecodingEv", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::enablelookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23generationLengthsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21generationLengthsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers25generationLengthsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::lookaheadruntimebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmaskhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers14packedMaskHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmaskhostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18packedMaskHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmasksdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17packedMasksDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17positionIdsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15positionIdsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionIdsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21positionOffsetsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionOffsetsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23positionOffsetsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::setfrominputs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorMapE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15useSpecDecodingE", false]], "tensorrt_llm::runtime::loracache (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCacheE", false]], "tensorrt_llm::runtime::loracache::bump (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::bumptaskinprogress (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::claimpageswithevict (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", false]], "tensorrt_llm::runtime::loracache::copytask (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", false]], "tensorrt_llm::runtime::loracache::copytaskmappages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", false]], "tensorrt_llm::runtime::loracache::copytopages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", false]], "tensorrt_llm::runtime::loracache::determinenumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", false], [1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", false]], "tensorrt_llm::runtime::loracache::fits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", false]], "tensorrt_llm::runtime::loracache::get (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::getnumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache11getNumPagesEv", false]], "tensorrt_llm::runtime::loracache::getpageptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", false]], "tensorrt_llm::runtime::loracache::getstatus (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::has (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::isdone (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::isloaded (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::loadweights (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::loracache::loracache (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::loracache::markalldone (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11markAllDoneEv", false]], "tensorrt_llm::runtime::loracache::marktaskdone (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::mbuffermanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache14mBufferManagerE", false]], "tensorrt_llm::runtime::loracache::mcachemap (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9mCacheMapE", false]], "tensorrt_llm::runtime::loracache::mcachemutex (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mCacheMutexE", false]], "tensorrt_llm::runtime::loracache::mcachepagemanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mCachePageManagerE", false]], "tensorrt_llm::runtime::loracache::mdevicebuffermanagers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21mDeviceBufferManagersE", false]], "tensorrt_llm::runtime::loracache::mdonetasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10mDoneTasksE", false]], "tensorrt_llm::runtime::loracache::minprogresstasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16mInProgressTasksE", false]], "tensorrt_llm::runtime::loracache::mmodelconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mModelConfigE", false]], "tensorrt_llm::runtime::loracache::mmoduleidtomodule (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mModuleIdToModuleE", false]], "tensorrt_llm::runtime::loracache::mpagemanagerconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18mPageManagerConfigE", false]], "tensorrt_llm::runtime::loracache::mpagesmutex (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mPagesMutexE", false]], "tensorrt_llm::runtime::loracache::mworldconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mWorldConfigE", false]], "tensorrt_llm::runtime::loracache::put (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", false]], "tensorrt_llm::runtime::loracache::splittransposecpu (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracache::splittransposecpuinner (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracache::taskidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10TaskIdTypeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::adaptersize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig11adapterSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::insize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6inSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::layerid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7layerIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::moduleid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8moduleIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::numslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8numSlotsE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::outsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7outSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::pageid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6pageIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::scalingvecpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17scalingVecPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::slotidx (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7slotIdxE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::tostring (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8toStringEv", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::weightsinpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig16weightsInPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::weightsoutpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17weightsOutPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfiglistptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache28TaskLayerModuleConfigListPtrE", false]], "tensorrt_llm::runtime::loracache::taskvalue (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueE", false]], "tensorrt_llm::runtime::loracache::taskvalue::configs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7configsE", false]], "tensorrt_llm::runtime::loracache::taskvalue::done (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue4doneE", false]], "tensorrt_llm::runtime::loracache::taskvalue::inprogress (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue10inProgressE", false]], "tensorrt_llm::runtime::loracache::taskvalue::it (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue2itE", false]], "tensorrt_llm::runtime::loracache::taskvalue::loaded (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue6loadedE", false]], "tensorrt_llm::runtime::loracache::taskvalue::loadinprogress (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue14loadInProgressE", false]], "tensorrt_llm::runtime::loracache::taskvalue::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", false]], "tensorrt_llm::runtime::loracache::taskvalue::pageids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7pageIdsE", false]], "tensorrt_llm::runtime::loracache::taskvalue::taskvalue (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueEv", false]], "tensorrt_llm::runtime::loracache::taskvalue::~taskvalue (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueD0Ev", false]], "tensorrt_llm::runtime::loracache::taskvalueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12TaskValuePtrE", false]], "tensorrt_llm::runtime::loracache::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TensorPtrE", false]], "tensorrt_llm::runtime::loracache::valuestatus (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatusE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_loaded (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_missing (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_processing (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", false]], "tensorrt_llm::runtime::loracachefullexception (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionE", false]], "tensorrt_llm::runtime::loracachefullexception::loracachefullexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", false]], "tensorrt_llm::runtime::loracachefullexception::~loracachefullexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionD0Ev", false]], "tensorrt_llm::runtime::loracachepagemanager (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManagerE", false]], "tensorrt_llm::runtime::loracachepagemanager::blockptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanager::claimpages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanager::initialize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", false]], "tensorrt_llm::runtime::loracachepagemanager::loracachepagemanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", false]], "tensorrt_llm::runtime::loracachepagemanager::mconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager7mConfigE", false]], "tensorrt_llm::runtime::loracachepagemanager::mfreepageids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12mFreePageIdsE", false]], "tensorrt_llm::runtime::loracachepagemanager::mispagefree (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mIsPageFreeE", false]], "tensorrt_llm::runtime::loracachepagemanager::mpageblocks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mPageBlocksE", false]], "tensorrt_llm::runtime::loracachepagemanager::mutablepageptr (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", false]], "tensorrt_llm::runtime::loracachepagemanager::numavailablepages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager17numAvailablePagesEv", false]], "tensorrt_llm::runtime::loracachepagemanager::pageptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", false]], "tensorrt_llm::runtime::loracachepagemanager::releasepages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", false]], "tensorrt_llm::runtime::loracachepagemanager::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager9TensorPtrE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfigE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig11getDataTypeEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getinittozero (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getInitToZeroEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getmaxpagesperblock (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig19getMaxPagesPerBlockEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getmemorytype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getMemoryTypeEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getnumcopystreams (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig17getNumCopyStreamsEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getpagewidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig12getPageWidthEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getslotsperpage (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig15getSlotsPerPageEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::gettotalnumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig16getTotalNumPagesEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::loracachepagemanagerconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig9mDataTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::minittozero (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mInitToZeroE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mmaxpagesperblock (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17mMaxPagesPerBlockE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mmemorytype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mMemoryTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mnumcopystreams (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15mNumCopyStreamsE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mpagewidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig10mPageWidthE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mslotsperpage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13mSlotsPerPageE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mtotalnumpages (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig14mTotalNumPagesE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setdatatype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setinittozero (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setmaxpagesperblock (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setmemorytype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setnumcopystreams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setpagewidth (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setslotsperpage (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::settotalnumpage (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", false]], "tensorrt_llm::runtime::loraexpectedexception (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionE", false]], "tensorrt_llm::runtime::loraexpectedexception::loraexpectedexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", false]], "tensorrt_llm::runtime::loraexpectedexception::~loraexpectedexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionD0Ev", false]], "tensorrt_llm::runtime::loramodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleE", false]], "tensorrt_llm::runtime::loramodule::createloramodules (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::flattenedinoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", false]], "tensorrt_llm::runtime::loramodule::indim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5inDimEv", false]], "tensorrt_llm::runtime::loramodule::indimfirst (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10inDimFirstEv", false]], "tensorrt_llm::runtime::loramodule::insize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::intpsplitdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12inTpSplitDimEv", false]], "tensorrt_llm::runtime::loramodule::localinadaptersize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localindim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::localinoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localinsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutadaptersize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localscalessize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", false]], "tensorrt_llm::runtime::loramodule::localtotalsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::loramodule::loramodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleEv", false]], "tensorrt_llm::runtime::loramodule::mindim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule6mInDimE", false]], "tensorrt_llm::runtime::loramodule::mindimfirst (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule11mInDimFirstE", false]], "tensorrt_llm::runtime::loramodule::mintpsplitdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule13mInTpSplitDimE", false]], "tensorrt_llm::runtime::loramodule::moduletype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleTypeE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_dense (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kATTN_DENSEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_k (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_KE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_q (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_QE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_qkv (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kATTN_QKVE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_v (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_VE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_dense (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType17kCROSS_ATTN_DENSEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_k (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_KE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_q (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_QE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_qkv (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType15kCROSS_ATTN_QKVE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_v (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_VE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kinvalid (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType8kINVALIDE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_4h_to_h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_4H_TO_HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_gate (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMLP_GATEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_gate_up (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_GATE_UPE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_h_to_4h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_H_TO_4HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_router (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMLP_ROUTERE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_4h_to_h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_4H_TO_HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_gate (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMOE_GATEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_h_to_4h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_H_TO_4HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_router (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMOE_ROUTERE", false]], "tensorrt_llm::runtime::loramodule::moutdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule7mOutDimE", false]], "tensorrt_llm::runtime::loramodule::moutdimfirst (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12mOutDimFirstE", false]], "tensorrt_llm::runtime::loramodule::mouttpsplitdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule14mOutTpSplitDimE", false]], "tensorrt_llm::runtime::loramodule::mtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule5mTypeE", false]], "tensorrt_llm::runtime::loramodule::name (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule4nameEv", false]], "tensorrt_llm::runtime::loramodule::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", false]], "tensorrt_llm::runtime::loramodule::outdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6outDimEv", false]], "tensorrt_llm::runtime::loramodule::outdimfirst (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11outDimFirstEv", false]], "tensorrt_llm::runtime::loramodule::outsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::outtpsplitdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule13outTpSplitDimEv", false]], "tensorrt_llm::runtime::loramodule::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule9TensorPtrE", false]], "tensorrt_llm::runtime::loramodule::tomodulename (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::tomoduletype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", false]], "tensorrt_llm::runtime::loramodule::value (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5valueEv", false]], "tensorrt_llm::runtime::lorataskidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14LoraTaskIdTypeE", false]], "tensorrt_llm::runtime::medusamodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModuleE", false]], "tensorrt_llm::runtime::medusamodule::getmedusachoices (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime12MedusaModule16getMedusaChoicesEv", false]], "tensorrt_llm::runtime::medusamodule::mdefaultmedusachoices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule21mDefaultMedusaChoicesE", false]], "tensorrt_llm::runtime::medusamodule::medusachoices (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule13MedusaChoicesE", false]], "tensorrt_llm::runtime::medusamodule::medusamodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleEv", false]], "tensorrt_llm::runtime::medusamodule::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule9TensorPtrE", false]], "tensorrt_llm::runtime::memorycounters (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCountersE", false]], "tensorrt_llm::runtime::memorycounters::allocate (c++ function)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", false]], "tensorrt_llm::runtime::memorycounters::bytestostring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", false]], "tensorrt_llm::runtime::memorycounters::deallocate (c++ function)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", false]], "tensorrt_llm::runtime::memorycounters::difftype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8DiffTypeE", false]], "tensorrt_llm::runtime::memorycounters::getcpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getCpuEv", false]], "tensorrt_llm::runtime::memorycounters::getcpudiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getCpuDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getgpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getGpuEv", false]], "tensorrt_llm::runtime::memorycounters::getgpudiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getGpuDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getinstance (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11getInstanceEv", false]], "tensorrt_llm::runtime::memorycounters::getpinned (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters9getPinnedEv", false]], "tensorrt_llm::runtime::memorycounters::getpinneddiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getpinnedpool (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedPoolEv", false]], "tensorrt_llm::runtime::memorycounters::getpinnedpooldiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters17getPinnedPoolDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getuvm (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getUVMEv", false]], "tensorrt_llm::runtime::memorycounters::getuvmdiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getUVMDiffEv", false]], "tensorrt_llm::runtime::memorycounters::mcpu (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mCpuE", false]], "tensorrt_llm::runtime::memorycounters::mcpudiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mCpuDiffE", false]], "tensorrt_llm::runtime::memorycounters::memorycounters (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters14MemoryCountersEv", false]], "tensorrt_llm::runtime::memorycounters::mgpu (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mGpuE", false]], "tensorrt_llm::runtime::memorycounters::mgpudiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mGpuDiffE", false]], "tensorrt_llm::runtime::memorycounters::mpinned (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters7mPinnedE", false]], "tensorrt_llm::runtime::memorycounters::mpinneddiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedDiffE", false]], "tensorrt_llm::runtime::memorycounters::mpinnedpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedPoolE", false]], "tensorrt_llm::runtime::memorycounters::mpinnedpooldiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters15mPinnedPoolDiffE", false]], "tensorrt_llm::runtime::memorycounters::muvm (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mUVME", false]], "tensorrt_llm::runtime::memorycounters::muvmdiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mUVMDiffE", false]], "tensorrt_llm::runtime::memorycounters::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10SizeType32E", false]], "tensorrt_llm::runtime::memorycounters::tostring (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters8toStringEv", false]], "tensorrt_llm::runtime::memorytype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryTypeE", false]], "tensorrt_llm::runtime::memorytype::kcpu (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kCPUE", false]], "tensorrt_llm::runtime::memorytype::kgpu (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kGPUE", false]], "tensorrt_llm::runtime::memorytype::kpinned (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType7kPINNEDE", false]], "tensorrt_llm::runtime::memorytype::kpinnedpool (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType11kPINNEDPOOLE", false]], "tensorrt_llm::runtime::memorytype::kuvm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kUVME", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEE5valueE", false]], "tensorrt_llm::runtime::modelconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::modelconfig::computecontextlogits (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEv", false]], "tensorrt_llm::runtime::modelconfig::computegenerationlogits (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEv", false]], "tensorrt_llm::runtime::modelconfig::countlocallayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::countlowerranklayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::disableseamlesslookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig32disableSeamlessLookaheadDecodingEv", false]], "tensorrt_llm::runtime::modelconfig::enableseamlesslookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getcontextfmha (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getContextFMHAEv", false]], "tensorrt_llm::runtime::modelconfig::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getDataTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getencoderhiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getEncoderHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getgemmallreducedtype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getGemmAllReduceDtypeEv", false]], "tensorrt_llm::runtime::modelconfig::gethiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getkvcachetype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getKVCacheTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getkvdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getKvDataTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getlayertypes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getLayerTypesEv", false]], "tensorrt_llm::runtime::modelconfig::getlogitsdtype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLogitsDtypeEv", false]], "tensorrt_llm::runtime::modelconfig::getloramodules (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLoraModulesEv", false]], "tensorrt_llm::runtime::modelconfig::getmanageweightstype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getManageWeightsTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxbatchsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBatchSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxdecodingdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig25getMaxDecodingDraftTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxdecodingtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getMaxDecodingTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxencoderlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMaxEncoderLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxinputlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxInputLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxlorarank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxLoraRankEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxnumtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxNumTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxpositionembeddings (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24getMaxPositionEmbeddingsEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxpromptembeddingtablesize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig30getMaxPromptEmbeddingTableSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxsequencelen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getMaxSequenceLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmlphiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMlpHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmodelname (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getModelNameEv", false]], "tensorrt_llm::runtime::modelconfig::getmodelvariant (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getModelVariantEv", false]], "tensorrt_llm::runtime::modelconfig::getnbattentionlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnbheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10getNbHeadsEv", false]], "tensorrt_llm::runtime::modelconfig::getnbkvheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnblayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnbrnnlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnumkvheadsperlayer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getNumKvHeadsPerLayerEv", false]], "tensorrt_llm::runtime::modelconfig::getnumkvheadsperlayerlocalrange (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::modelconfig::getnumlanguages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getNumLanguagesEv", false]], "tensorrt_llm::runtime::modelconfig::getoptprofilessplitpoints (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig25getOptProfilesSplitPointsEv", false]], "tensorrt_llm::runtime::modelconfig::getpagedcontextfmha (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19getPagedContextFMHAEv", false]], "tensorrt_llm::runtime::modelconfig::getppreducescatter (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getPpReduceScatterEv", false]], "tensorrt_llm::runtime::modelconfig::getquantmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getQuantModeEv", false]], "tensorrt_llm::runtime::modelconfig::getrnnconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getRnnConfigEv", false]], "tensorrt_llm::runtime::modelconfig::getrotaryembeddingdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getRotaryEmbeddingDimEv", false]], "tensorrt_llm::runtime::modelconfig::getsizeperhead (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getSizePerHeadEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig26getSpeculativeDecodingModeEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmodule (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28getSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmoduleptr (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", false]], "tensorrt_llm::runtime::modelconfig::getsumlocalkvheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::modelconfig::gettokensperblock (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getTokensPerBlockEv", false]], "tensorrt_llm::runtime::modelconfig::getvocabsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getVocabSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getvocabsizepadded (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::hasrnnconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12hasRnnConfigEv", false]], "tensorrt_llm::runtime::modelconfig::hasspeculativedecodingmodule (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28hasSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::iscontinuouskvcache (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19isContinuousKVCacheEv", false]], "tensorrt_llm::runtime::modelconfig::iskvcacheenabled (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16isKVCacheEnabledEv", false]], "tensorrt_llm::runtime::modelconfig::ismultimodal (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12isMultiModalEv", false]], "tensorrt_llm::runtime::modelconfig::ispagedkvcache (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14isPagedKVCacheEv", false]], "tensorrt_llm::runtime::modelconfig::isrnnbased (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10isRnnBasedEv", false]], "tensorrt_llm::runtime::modelconfig::istransformerbased (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18isTransformerBasedEv", false]], "tensorrt_llm::runtime::modelconfig::iswhisper (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig9isWhisperEv", false]], "tensorrt_llm::runtime::modelconfig::kdefault_num_tokens_per_block (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig29kDEFAULT_NUM_TOKENS_PER_BLOCKE", false]], "tensorrt_llm::runtime::modelconfig::kopt_profiles_split_points (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26kOPT_PROFILES_SPLIT_POINTSE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheTypeE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kcontinuous (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType11kCONTINUOUSE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kdisabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType9kDISABLEDE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kpaged (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType6kPAGEDE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetypefromstring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", false]], "tensorrt_llm::runtime::modelconfig::layertype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerTypeE", false]], "tensorrt_llm::runtime::modelconfig::layertype::kattention (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kATTENTIONE", false]], "tensorrt_llm::runtime::modelconfig::layertype::klinear (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType7kLINEARE", false]], "tensorrt_llm::runtime::modelconfig::layertype::knoop (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType5kNOOPE", false]], "tensorrt_llm::runtime::modelconfig::layertype::krecurrent (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kRECURRENTE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsTypeE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype::kdisabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType9kDisabledE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype::kenabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType8kEnabledE", false]], "tensorrt_llm::runtime::modelconfig::mcomputecontextlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mComputeContextLogitsE", false]], "tensorrt_llm::runtime::modelconfig::mcomputegenerationlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mComputeGenerationLogitsE", false]], "tensorrt_llm::runtime::modelconfig::mcontextfmha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mContextFMHAE", false]], "tensorrt_llm::runtime::modelconfig::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mDataTypeE", false]], "tensorrt_llm::runtime::modelconfig::mencoderhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mEncoderHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::mgemmallreducedtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mGemmAllReduceDtypeE", false]], "tensorrt_llm::runtime::modelconfig::mhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::minputpacked (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mInputPackedE", false]], "tensorrt_llm::runtime::modelconfig::mkvcachetype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mKVCacheTypeE", false]], "tensorrt_llm::runtime::modelconfig::mlayertypes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mLayerTypesE", false]], "tensorrt_llm::runtime::modelconfig::mlogitsdtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLogitsDtypeE", false]], "tensorrt_llm::runtime::modelconfig::mloramodules (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLoraModulesE", false]], "tensorrt_llm::runtime::modelconfig::mmanageweightstype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mManageWeightsTypeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxbeamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBeamWidthE", false]], "tensorrt_llm::runtime::modelconfig::mmaxencoderlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMaxEncoderLenE", false]], "tensorrt_llm::runtime::modelconfig::mmaxinputlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxInputLenE", false]], "tensorrt_llm::runtime::modelconfig::mmaxlorarank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxLoraRankE", false]], "tensorrt_llm::runtime::modelconfig::mmaxnumtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxNumTokensE", false]], "tensorrt_llm::runtime::modelconfig::mmaxpositionembeddings (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mMaxPositionEmbeddingsE", false]], "tensorrt_llm::runtime::modelconfig::mmaxpromptembeddingtablesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mMaxPromptEmbeddingTableSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxsequencelen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mMaxSequenceLenE", false]], "tensorrt_llm::runtime::modelconfig::mmlphiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMlpHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmodelname (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mModelNameE", false]], "tensorrt_llm::runtime::modelconfig::mmodelvariant (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mModelVariantE", false]], "tensorrt_llm::runtime::modelconfig::mnbattentionlayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mNbAttentionLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnbheads (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig8mNbHeadsE", false]], "tensorrt_llm::runtime::modelconfig::mnblayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mNbLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnbrnnlayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mNbRnnLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnumkvheadsperattentionlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mNumKvHeadsPerAttentionLayerE", false]], "tensorrt_llm::runtime::modelconfig::mnumkvheadspercrossattentionlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig33mNumKvHeadsPerCrossAttentionLayerE", false]], "tensorrt_llm::runtime::modelconfig::mnumlanguages (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mNumLanguagesE", false]], "tensorrt_llm::runtime::modelconfig::modelconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariantE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kchatglm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant8kChatGlmE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kencdec (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant7kEncDecE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kglm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGlmE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kgpt (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGptE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kmamba (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant6kMambaE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::krecurrentgemma (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant15kRecurrentGemmaE", false]], "tensorrt_llm::runtime::modelconfig::mpagedcontextfmha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17mPagedContextFMHAE", false]], "tensorrt_llm::runtime::modelconfig::mpagedstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mPagedStateE", false]], "tensorrt_llm::runtime::modelconfig::mppreducescatter (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16mPpReduceScatterE", false]], "tensorrt_llm::runtime::modelconfig::mquantmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mQuantModeE", false]], "tensorrt_llm::runtime::modelconfig::mrnnconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mRnnConfigE", false]], "tensorrt_llm::runtime::modelconfig::mrotaryembeddingdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mRotaryEmbeddingDimE", false]], "tensorrt_llm::runtime::modelconfig::msizeperhead (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mSizePerHeadE", false]], "tensorrt_llm::runtime::modelconfig::mskipcrossattnblocks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20mSkipCrossAttnBlocksE", false]], "tensorrt_llm::runtime::modelconfig::mspeculativedecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mSpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::modelconfig::mspeculativedecodingmodule (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26mSpeculativeDecodingModuleE", false]], "tensorrt_llm::runtime::modelconfig::mtokensperblock (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mTokensPerBlockE", false]], "tensorrt_llm::runtime::modelconfig::musecrossattention (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseCrossAttentionE", false]], "tensorrt_llm::runtime::modelconfig::musegemmallreduceplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23mUseGemmAllReducePluginE", false]], "tensorrt_llm::runtime::modelconfig::musegptattentionplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseGptAttentionPluginE", false]], "tensorrt_llm::runtime::modelconfig::museloraplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mUseLoraPluginE", false]], "tensorrt_llm::runtime::modelconfig::musemambaconv1dplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUseMambaConv1dPluginE", false]], "tensorrt_llm::runtime::modelconfig::musemrope (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mUseMropeE", false]], "tensorrt_llm::runtime::modelconfig::musepositionembedding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUsePositionEmbeddingE", false]], "tensorrt_llm::runtime::modelconfig::museshapeinference (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseShapeInferenceE", false]], "tensorrt_llm::runtime::modelconfig::musetokentypeembedding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseTokenTypeEmbeddingE", false]], "tensorrt_llm::runtime::modelconfig::mvocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mVocabSizeE", false]], "tensorrt_llm::runtime::modelconfig::resetspeculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30resetSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfigE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::convkernel (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig10convKernelE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnconvdimsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig14rnnConvDimSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnheadsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig11rnnHeadSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig13rnnHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::statesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig9stateSizeE", false]], "tensorrt_llm::runtime::modelconfig::setcontextfmha (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", false]], "tensorrt_llm::runtime::modelconfig::setencoderhiddensize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setgemmallreducedtype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::setkvcachetype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", false]], "tensorrt_llm::runtime::modelconfig::setlayertypes (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", false]], "tensorrt_llm::runtime::modelconfig::setlogitsdtype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::setloramodules (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", false]], "tensorrt_llm::runtime::modelconfig::setmanageweightstype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", false]], "tensorrt_llm::runtime::modelconfig::setmaxbatchsize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxbeamwidth (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxencoderlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxinputlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxlorarank (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxnumtokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setmaxpositionembeddings (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxpromptembeddingtablesize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxsequencelen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmlphiddensize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmodelname (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", false]], "tensorrt_llm::runtime::modelconfig::setmodelvariant (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", false]], "tensorrt_llm::runtime::modelconfig::setnbcrosskvheads (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setnbkvheads (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setnumkvheadspercrosslayer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setnumkvheadsperlayer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setnumlanguages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setpagedcontextfmha (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", false]], "tensorrt_llm::runtime::modelconfig::setppreducescatter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", false]], "tensorrt_llm::runtime::modelconfig::setquantmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", false]], "tensorrt_llm::runtime::modelconfig::setrnnconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", false]], "tensorrt_llm::runtime::modelconfig::setrotaryembeddingdim (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setsizeperhead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setskipcrossattnblocks (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", false]], "tensorrt_llm::runtime::modelconfig::setspeculativedecodingmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", false]], "tensorrt_llm::runtime::modelconfig::setspeculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::modelconfig::settokensperblock (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setusecrossattention (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", false]], "tensorrt_llm::runtime::modelconfig::setusemrope (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", false]], "tensorrt_llm::runtime::modelconfig::setusepositionembedding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", false]], "tensorrt_llm::runtime::modelconfig::setuseshapeinference (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", false]], "tensorrt_llm::runtime::modelconfig::setusetokentypeembedding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", false]], "tensorrt_llm::runtime::modelconfig::skipcrossattnblocks (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19skipCrossAttnBlocksEv", false]], "tensorrt_llm::runtime::modelconfig::supportsinflightbatching (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24supportsInflightBatchingEv", false]], "tensorrt_llm::runtime::modelconfig::usecrossattention (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useCrossAttentionEv", false]], "tensorrt_llm::runtime::modelconfig::usegemmallreduceplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEv", false]], "tensorrt_llm::runtime::modelconfig::usegptattentionplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEv", false]], "tensorrt_llm::runtime::modelconfig::uselanguageadapter (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18useLanguageAdapterEv", false]], "tensorrt_llm::runtime::modelconfig::useloraplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13useLoraPluginEv", false]], "tensorrt_llm::runtime::modelconfig::usemambaconv1dplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEv", false]], "tensorrt_llm::runtime::modelconfig::usemrope (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig8useMropeEv", false]], "tensorrt_llm::runtime::modelconfig::usepackedinput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14usePackedInputEv", false]], "tensorrt_llm::runtime::modelconfig::usepagedstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13usePagedStateEv", false]], "tensorrt_llm::runtime::modelconfig::usepositionembedding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20usePositionEmbeddingEv", false]], "tensorrt_llm::runtime::modelconfig::useprompttuning (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15usePromptTuningEv", false]], "tensorrt_llm::runtime::modelconfig::useshapeinference (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useShapeInferenceEv", false]], "tensorrt_llm::runtime::modelconfig::usetokentypeembedding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useTokenTypeEmbeddingEv", false]], "tensorrt_llm::runtime::mpi_group_barrier (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", false]], "tensorrt_llm::runtime::operator<< (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::pointerelementtype (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", false]], "tensorrt_llm::runtime::prompttuningparams (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParamsE", false]], "tensorrt_llm::runtime::prompttuningparams::filltaskstensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", false]], "tensorrt_llm::runtime::prompttuningparams::prompttuningparams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::prompttuningparams::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams10SizeType32E", false]], "tensorrt_llm::runtime::prompttuningparams::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams9TensorPtrE", false]], "tensorrt_llm::runtime::rawengine (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngineE", false]], "tensorrt_llm::runtime::rawengine::getaddress (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getAddressEv", false]], "tensorrt_llm::runtime::rawengine::gethostmemory (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine13getHostMemoryEv", false]], "tensorrt_llm::runtime::rawengine::getmanagedweightsmapopt (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine23getManagedWeightsMapOptEv", false]], "tensorrt_llm::runtime::rawengine::getpath (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getPathEv", false]], "tensorrt_llm::runtime::rawengine::getpathopt (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getPathOptEv", false]], "tensorrt_llm::runtime::rawengine::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getSizeEv", false]], "tensorrt_llm::runtime::rawengine::gettype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getTypeEv", false]], "tensorrt_llm::runtime::rawengine::mengineaddr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineAddrE", false]], "tensorrt_llm::runtime::rawengine::menginebuffer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine13mEngineBufferE", false]], "tensorrt_llm::runtime::rawengine::menginepath (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEnginePathE", false]], "tensorrt_llm::runtime::rawengine::menginesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineSizeE", false]], "tensorrt_llm::runtime::rawengine::mmanagedweightsmap (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine18mManagedWeightsMapE", false]], "tensorrt_llm::runtime::rawengine::mtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine5mTypeE", false]], "tensorrt_llm::runtime::rawengine::rawengine (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", false], [1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", false], [1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", false]], "tensorrt_llm::runtime::rawengine::setmanagedweightsmap (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", false]], "tensorrt_llm::runtime::rawengine::setpath (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", false]], "tensorrt_llm::runtime::rawengine::type (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4TypeE", false]], "tensorrt_llm::runtime::rawengine::type::addresswithsize (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", false]], "tensorrt_llm::runtime::rawengine::type::filepath (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", false]], "tensorrt_llm::runtime::rawengine::type::hostmemory (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", false]], "tensorrt_llm::runtime::requesttype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestTypeE", false]], "tensorrt_llm::runtime::requesttype::kcontext (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestType8kCONTEXTE", false]], "tensorrt_llm::runtime::requesttype::kgeneration (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestType11kGENERATIONE", false]], "tensorrt_llm::runtime::runtimedefaults (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaultsE", false]], "tensorrt_llm::runtime::runtimedefaults::maxattentionwindowvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults21maxAttentionWindowVecE", false]], "tensorrt_llm::runtime::runtimedefaults::runtimedefaults (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", false], [1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsEv", false]], "tensorrt_llm::runtime::runtimedefaults::sinktokenlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15sinkTokenLengthE", false]], "tensorrt_llm::runtime::samplingconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfigE", false]], "tensorrt_llm::runtime::samplingconfig::beamsearchdiversityrate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig23beamSearchDiversityRateE", false]], "tensorrt_llm::runtime::samplingconfig::beamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9beamWidthE", false]], "tensorrt_llm::runtime::samplingconfig::beamwidtharray (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14beamWidthArrayE", false]], "tensorrt_llm::runtime::samplingconfig::cumlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11cumLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::draftacceptancethreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig24draftAcceptanceThresholdE", false]], "tensorrt_llm::runtime::samplingconfig::earlystopping (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13earlyStoppingE", false]], "tensorrt_llm::runtime::samplingconfig::floattype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9FloatTypeE", false]], "tensorrt_llm::runtime::samplingconfig::frequencypenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig16frequencyPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::fusevalues (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", false]], "tensorrt_llm::runtime::samplingconfig::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::samplingconfig::getnumreturnbeams (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig17getNumReturnBeamsEv", false]], "tensorrt_llm::runtime::samplingconfig::lengthpenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13lengthPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::minlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9minLengthE", false]], "tensorrt_llm::runtime::samplingconfig::minp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4minPE", false]], "tensorrt_llm::runtime::samplingconfig::norepeatngramsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17noRepeatNgramSizeE", false]], "tensorrt_llm::runtime::samplingconfig::normalizelogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17normalizeLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::numreturnsequences (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig18numReturnSequencesE", false]], "tensorrt_llm::runtime::samplingconfig::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", false]], "tensorrt_llm::runtime::samplingconfig::optvec (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", false]], "tensorrt_llm::runtime::samplingconfig::originaltemperature (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig19originalTemperatureE", false]], "tensorrt_llm::runtime::samplingconfig::outputlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14outputLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::presencepenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15presencePenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::randomseed (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig10randomSeedE", false]], "tensorrt_llm::runtime::samplingconfig::repetitionpenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17repetitionPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::samplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", false], [1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", false]], "tensorrt_llm::runtime::samplingconfig::temperature (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11temperatureE", false]], "tensorrt_llm::runtime::samplingconfig::topk (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topKE", false]], "tensorrt_llm::runtime::samplingconfig::topkmedusaheads (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15topKMedusaHeadsE", false]], "tensorrt_llm::runtime::samplingconfig::topp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topPE", false]], "tensorrt_llm::runtime::samplingconfig::toppdecay (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9topPDecayE", false]], "tensorrt_llm::runtime::samplingconfig::toppmin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig7topPMinE", false]], "tensorrt_llm::runtime::samplingconfig::toppresetids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig12topPResetIdsE", false]], "tensorrt_llm::runtime::samplingconfig::usedefaultvalues (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", false]], "tensorrt_llm::runtime::samplingconfig::validate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig8validateEv", false]], "tensorrt_llm::runtime::samplingconfig::validatevec (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", false]], "tensorrt_llm::runtime::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10SizeType32E", false]], "tensorrt_llm::runtime::sizetype64 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10SizeType64E", false]], "tensorrt_llm::runtime::speculativedecodingmode (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::speculativedecodingmode::allbitset (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::anybitset (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::drafttokensexternal (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19DraftTokensExternalEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::eagle (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5EagleEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::explicitdrafttokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19ExplicitDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::hasdraftlogits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode14hasDraftLogitsEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isdrafttokensexternal (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isDraftTokensExternalEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::iseagle (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode7isEagleEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isexplicitdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isExplicitDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::islookaheaddecoding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19isLookaheadDecodingEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::ismedusa (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode8isMedusaEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isnone (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode6isNoneEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::kdrafttokensexternal (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kDraftTokensExternalE", false]], "tensorrt_llm::runtime::speculativedecodingmode::keagle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6kEagleE", false]], "tensorrt_llm::runtime::speculativedecodingmode::kexplicitdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kExplicitDraftTokensE", false]], "tensorrt_llm::runtime::speculativedecodingmode::klookaheaddecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode18kLookaheadDecodingE", false]], "tensorrt_llm::runtime::speculativedecodingmode::kmedusa (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode7kMedusaE", false]], "tensorrt_llm::runtime::speculativedecodingmode::knone (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5kNoneE", false]], "tensorrt_llm::runtime::speculativedecodingmode::lookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode17LookaheadDecodingEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::medusa (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6MedusaEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::mstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6mStateE", false]], "tensorrt_llm::runtime::speculativedecodingmode::needsdecoderprologue (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode20needsDecoderPrologueEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::needskvcacherewind (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18needsKVCacheRewindEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::none (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode4NoneEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", false]], "tensorrt_llm::runtime::speculativedecodingmode::predictsdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19predictsDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::requiresattentionmask (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21requiresAttentionMaskEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::speculativedecodingmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::underlyingtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode14UnderlyingTypeE", false]], "tensorrt_llm::runtime::speculativedecodingmode::updatespositionids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18updatesPositionIdsEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::variabledraftlength (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19variableDraftLengthEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::computenumpackedmasks (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule21computeNumPackedMasksEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdecodingdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule25getMaxDecodingDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdecodingtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule20getMaxDecodingTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdraftpathlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule18getMaxDraftPathLenEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxnumpaths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule14getMaxNumPathsEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxpathlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule13getMaxPathLenEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getnumpackedmasks (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule17getNumPackedMasksEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxdecodingdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule23mMaxDecodingDraftTokensE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxdraftpathlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule16mMaxDraftPathLenE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxnumpackedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18mMaxNumPackedMasksE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxnumpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule12mMaxNumPathsE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxdraftpathlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxdrafttokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxnumpaths (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::speculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", false], [1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::~speculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleD0Ev", false]], "tensorrt_llm::runtime::stringptrmap (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", false]], "tensorrt_llm::runtime::tllmlogger (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLoggerE", false]], "tensorrt_llm::runtime::tllmlogger::getlevel (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8getLevelEv", false]], "tensorrt_llm::runtime::tllmlogger::log (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", false]], "tensorrt_llm::runtime::tllmlogger::setlevel (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", false]], "tensorrt_llm::runtime::to_string (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", false], [1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::tokenextraidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16TokenExtraIdTypeE", false]], "tensorrt_llm::runtime::tokenidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11TokenIdTypeE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIbEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIbE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIfEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIfE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeI4halfEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeI4halfE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", false]], "tensorrt_llm::runtime::trtdatatype::kunderlyingtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE15kUnderlyingTypeE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIPvEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIPvE5valueE", false]], "tensorrt_llm::runtime::uniquetoken (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueTokenE", false]], "tensorrt_llm::runtime::uniquetoken::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", false]], "tensorrt_llm::runtime::uniquetoken::tokenextraid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken12tokenExtraIdE", false]], "tensorrt_llm::runtime::uniquetoken::tokenid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken7tokenIdE", false]], "tensorrt_llm::runtime::vectokenextraids (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16VecTokenExtraIdsE", false]], "tensorrt_llm::runtime::vecuniquetokens (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime15VecUniqueTokensE", false]], "tensorrt_llm::runtime::worldconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::worldconfig::enableattentiondp (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17enableAttentionDPEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getContextParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getContextParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getContextParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::getdevice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig9getDeviceEv", false]], "tensorrt_llm::runtime::worldconfig::getdeviceof (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", false]], "tensorrt_llm::runtime::worldconfig::getgpuspergroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig15getGpusPerGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getgpuspernode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14getGpusPerNodeEv", false]], "tensorrt_llm::runtime::worldconfig::getlastrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getLastRankEv", false]], "tensorrt_llm::runtime::worldconfig::getlocalrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig12getLocalRankEv", false]], "tensorrt_llm::runtime::worldconfig::getnoderank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getNodeRankEv", false]], "tensorrt_llm::runtime::worldconfig::getnoderankof (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig24getPipelineParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getPipelineParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::getrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getRankEv", false]], "tensorrt_llm::runtime::worldconfig::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getSizeEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getTensorParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig20getTensorParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getTensorParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::iscontextparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17isContextParallelEv", false]], "tensorrt_llm::runtime::worldconfig::isfirstcontextparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isFirstContextParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::isfirstpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig27isFirstPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::isfirsttensorparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig25isFirstTensorParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::islastpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isLastPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::ispipelineparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig18isPipelineParallelEv", false]], "tensorrt_llm::runtime::worldconfig::istensorparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig16isTensorParallelEv", false]], "tensorrt_llm::runtime::worldconfig::kdefaultgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19kDefaultGpusPerNodeE", false]], "tensorrt_llm::runtime::worldconfig::mcontextparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19mContextParallelismE", false]], "tensorrt_llm::runtime::worldconfig::mdeviceids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig10mDeviceIdsE", false]], "tensorrt_llm::runtime::worldconfig::menableattentiondp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mEnableAttentionDPE", false]], "tensorrt_llm::runtime::worldconfig::mgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig12mGpusPerNodeE", false]], "tensorrt_llm::runtime::worldconfig::mpi (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", false]], "tensorrt_llm::runtime::worldconfig::mpipelineparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig20mPipelineParallelismE", false]], "tensorrt_llm::runtime::worldconfig::mrank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig5mRankE", false]], "tensorrt_llm::runtime::worldconfig::mtensorparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mTensorParallelismE", false]], "tensorrt_llm::runtime::worldconfig::validmpiconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14validMpiConfigEv", false]], "tensorrt_llm::runtime::worldconfig::worldconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", false]], "text (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.text", false]], "text_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.text_diff", false]], "text_diff (tensorrt_llm.llmapi.completionoutput property)": [[65, "id4", false]], "timestepembedding (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.TimestepEmbedding", false]], "timesteps (class in tensorrt_llm.layers.embedding)": [[78, "tensorrt_llm.layers.embedding.Timesteps", false]], "to_dict() (tensorrt_llm.llmapi.buildconfig method)": [[65, "tensorrt_llm.llmapi.BuildConfig.to_dict", false]], "to_dict() (tensorrt_llm.llmapi.calibconfig method)": [[65, "tensorrt_llm.llmapi.CalibConfig.to_dict", false]], "to_dict() (tensorrt_llm.llmapi.quantconfig method)": [[65, "tensorrt_llm.llmapi.QuantConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.chatglmconfig method)": [[79, "tensorrt_llm.models.ChatGLMConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.cogvlmconfig method)": [[79, "tensorrt_llm.models.CogVLMConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.dbrxconfig method)": [[79, "tensorrt_llm.models.DbrxConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.falconconfig method)": [[79, "tensorrt_llm.models.FalconConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gemmaconfig method)": [[79, "tensorrt_llm.models.GemmaConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gptconfig method)": [[79, "tensorrt_llm.models.GPTConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gptjconfig method)": [[79, "tensorrt_llm.models.GPTJConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.llamaconfig method)": [[79, "tensorrt_llm.models.LLaMAConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.medusaconfig method)": [[79, "tensorrt_llm.models.MedusaConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.to_dict", false]], "to_json_file() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.to_json_file", false]], "to_layer_quant_config() (tensorrt_llm.models.pretrainedconfig method)": [[79, "tensorrt_llm.models.PretrainedConfig.to_layer_quant_config", false]], "to_legacy_setting() (tensorrt_llm.plugin.pluginconfig method)": [[80, "tensorrt_llm.plugin.PluginConfig.to_legacy_setting", false]], "token_drop() (tensorrt_llm.layers.embedding.labelembedding method)": [[78, "tensorrt_llm.layers.embedding.LabelEmbedding.token_drop", false]], "token_end (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.token_end", false]], "token_ids (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.token_ids", false]], "token_ids_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[65, "tensorrt_llm.llmapi.CompletionOutput.token_ids_diff", false]], "token_ids_diff (tensorrt_llm.llmapi.completionoutput property)": [[65, "id5", false]], "token_range_retention_configs (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.token_range_retention_configs", false]], "token_start (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[65, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.token_start", false]], "tokenizer (tensorrt_llm.llmapi.llm attribute)": [[65, "tensorrt_llm.llmapi.LLM.tokenizer", false]], "tokenizer (tensorrt_llm.llmapi.llm property)": [[65, "id0", false]], "tokenizer_image_token() (tensorrt_llm.runtime.multimodalmodelrunner static method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.tokenizer_image_token", false]], "tokenizer_max_seq_length (tensorrt_llm.llmapi.calibconfig attribute)": [[65, "tensorrt_llm.llmapi.CalibConfig.tokenizer_max_seq_length", false]], "tokens_per_block (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.tokens_per_block", false]], "tokens_per_block (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.tokens_per_block", false]], "top_k (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.top_k", false]], "top_k (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.top_k", false]], "top_p (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.top_p", false]], "top_p (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.top_p", false]], "top_p_decay (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.top_p_decay", false]], "top_p_decay (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.top_p_decay", false]], "top_p_min (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.top_p_min", false]], "top_p_min (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.top_p_min", false]], "top_p_reset_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.top_p_reset_ids", false]], "top_p_reset_ids (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.top_p_reset_ids", false]], "topk() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.topk", false]], "tp_split_dim() (tensorrt_llm.layers.linear.linear class method)": [[78, "tensorrt_llm.layers.linear.Linear.tp_split_dim", false]], "tp_split_dim() (tensorrt_llm.layers.linear.linearbase class method)": [[78, "tensorrt_llm.layers.linear.LinearBase.tp_split_dim", false]], "tp_split_dim() (tensorrt_llm.layers.linear.rowlinear class method)": [[78, "tensorrt_llm.layers.linear.RowLinear.tp_split_dim", false]], "transpose() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.transpose", false]], "transpose() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.transpose", false]], "trtllm-serve-disaggregated command line option": [[26, "cmdoption-trtllm-serve-disaggregated-c", false], [26, "cmdoption-trtllm-serve-disaggregated-r", false], [26, "cmdoption-trtllm-serve-disaggregated-t", false]], "trtllm-serve-disaggregated_mpi_worker command line option": [[26, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false], [26, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", false]], "trtllm-serve-serve command line option": [[26, "cmdoption-trtllm-serve-serve-arg-MODEL", false], [26, "cmdoption-trtllm-serve-serve-backend", false], [26, "cmdoption-trtllm-serve-serve-cluster_size", false], [26, "cmdoption-trtllm-serve-serve-ep_size", false], [26, "cmdoption-trtllm-serve-serve-extra_llm_api_options", false], [26, "cmdoption-trtllm-serve-serve-gpus_per_node", false], [26, "cmdoption-trtllm-serve-serve-host", false], [26, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", false], [26, "cmdoption-trtllm-serve-serve-log_level", false], [26, "cmdoption-trtllm-serve-serve-max_batch_size", false], [26, "cmdoption-trtllm-serve-serve-max_beam_width", false], [26, "cmdoption-trtllm-serve-serve-max_num_tokens", false], [26, "cmdoption-trtllm-serve-serve-max_seq_len", false], [26, "cmdoption-trtllm-serve-serve-num_postprocess_workers", false], [26, "cmdoption-trtllm-serve-serve-port", false], [26, "cmdoption-trtllm-serve-serve-pp_size", false], [26, "cmdoption-trtllm-serve-serve-reasoning_parser", false], [26, "cmdoption-trtllm-serve-serve-tokenizer", false], [26, "cmdoption-trtllm-serve-serve-tp_size", false], [26, "cmdoption-trtllm-serve-serve-trust_remote_code", false]], "trtllm_modules_to_hf_modules (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.trtllm_modules_to_hf_modules", false]], "truncate_prompt_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.truncate_prompt_tokens", false]], "twoshot (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.TWOSHOT", false]], "ub (tensorrt_llm.functional.allreducestrategy attribute)": [[77, "tensorrt_llm.functional.AllReduceStrategy.UB", false]], "unary() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.unary", false]], "unbind() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.unbind", false]], "unbind() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.unbind", false]], "unfuse_qkv_projections() (tensorrt_llm.models.sd3transformer2dmodel method)": [[79, "tensorrt_llm.models.SD3Transformer2DModel.unfuse_qkv_projections", false]], "unpatchify() (tensorrt_llm.models.dit method)": [[79, "tensorrt_llm.models.DiT.unpatchify", false]], "unsqueeze() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.unsqueeze", false]], "unsqueeze() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.unsqueeze", false]], "update() (tensorrt_llm.llmapi.buildconfig method)": [[65, "tensorrt_llm.llmapi.BuildConfig.update", false]], "update() (tensorrt_llm.runtime.samplingconfig method)": [[82, "tensorrt_llm.runtime.SamplingConfig.update", false]], "update_from_dict() (tensorrt_llm.llmapi.buildconfig method)": [[65, "tensorrt_llm.llmapi.BuildConfig.update_from_dict", false]], "update_kv_cache_type() (tensorrt_llm.llmapi.buildconfig method)": [[65, "tensorrt_llm.llmapi.BuildConfig.update_kv_cache_type", false]], "update_output_ids_by_offset() (tensorrt_llm.runtime.generationsession method)": [[82, "tensorrt_llm.runtime.GenerationSession.update_output_ids_by_offset", false]], "update_strategy() (tensorrt_llm.functional.allreduceparams method)": [[77, "tensorrt_llm.functional.AllReduceParams.update_strategy", false]], "use_beam_hyps (tensorrt_llm.runtime.samplingconfig attribute)": [[82, "tensorrt_llm.runtime.SamplingConfig.use_beam_hyps", false]], "use_beam_search (tensorrt_llm.llmapi.samplingparams attribute)": [[65, "tensorrt_llm.llmapi.SamplingParams.use_beam_search", false]], "use_dynamic_tree (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.EagleDecodingConfig.use_dynamic_tree", false]], "use_gemm_allreduce_plugin (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.use_gemm_allreduce_plugin", false]], "use_gpt_attention_plugin (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.use_gpt_attention_plugin", false]], "use_kv_cache (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.use_kv_cache", false]], "use_lora() (tensorrt_llm.models.decodermodel method)": [[79, "tensorrt_llm.models.DecoderModel.use_lora", false]], "use_lora() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.use_lora", false]], "use_lora() (tensorrt_llm.models.gemmaforcausallm method)": [[79, "tensorrt_llm.models.GemmaForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.gptforcausallm method)": [[79, "tensorrt_llm.models.GPTForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.llamaforcausallm method)": [[79, "tensorrt_llm.models.LLaMAForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.mllamaforcausallm method)": [[79, "tensorrt_llm.models.MLLaMAForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.phi3forcausallm method)": [[79, "tensorrt_llm.models.Phi3ForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.phiforcausallm method)": [[79, "tensorrt_llm.models.PhiForCausalLM.use_lora", false]], "use_lora_plugin (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.use_lora_plugin", false]], "use_lora_plugin (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.use_lora_plugin", false]], "use_mamba_conv1d_plugin (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.use_mamba_conv1d_plugin", false]], "use_meta_recipe (tensorrt_llm.llmapi.quantconfig attribute)": [[65, "tensorrt_llm.llmapi.QuantConfig.use_meta_recipe", false]], "use_mrope (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.use_mrope", false]], "use_prompt_tuning() (tensorrt_llm.models.encodermodel method)": [[79, "tensorrt_llm.models.EncoderModel.use_prompt_tuning", false]], "use_refit (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.use_refit", false]], "use_relaxed_acceptance_for_thinking (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[65, "tensorrt_llm.llmapi.MTPDecodingConfig.use_relaxed_acceptance_for_thinking", false]], "use_strip_plan (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.use_strip_plan", false]], "validate_positive_values() (tensorrt_llm.llmapi.lookaheaddecodingconfig class method)": [[65, "tensorrt_llm.llmapi.LookaheadDecodingConfig.validate_positive_values", false]], "verbatim (tensorrt_llm.models.gemmaconfig attribute)": [[79, "tensorrt_llm.models.GemmaConfig.VERBATIM", false]], "video_preprocess() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.video_preprocess", false]], "view() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.view", false]], "view() (tensorrt_llm.functional.tensor method)": [[77, "tensorrt_llm.functional.Tensor.view", false]], "view() (tensorrt_llm.runtime.tensorinfo method)": [[82, "tensorrt_llm.runtime.TensorInfo.view", false]], "visual_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[82, "tensorrt_llm.runtime.MultimodalModelRunner.visual_engine_dir", false]], "visualize_network (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.visualize_network", false]], "vocab_size (tensorrt_llm.runtime.generationsession property)": [[82, "tensorrt_llm.runtime.GenerationSession.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelconfig attribute)": [[82, "tensorrt_llm.runtime.ModelConfig.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.vocab_size", false]], "vocab_size_padded (tensorrt_llm.runtime.modelrunner property)": [[82, "tensorrt_llm.runtime.ModelRunner.vocab_size_padded", false]], "vocab_size_padded (tensorrt_llm.runtime.modelrunnercpp property)": [[82, "tensorrt_llm.runtime.ModelRunnerCpp.vocab_size_padded", false]], "w4a16 (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A16", false]], "w4a16_awq (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A16_AWQ", false]], "w4a16_gptq (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A16_GPTQ", false]], "w4a8_awq (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A8_AWQ", false]], "w4a8_qserve_per_channel (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A8_QSERVE_PER_CHANNEL", false]], "w4a8_qserve_per_group (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W4A8_QSERVE_PER_GROUP", false]], "w8a16 (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A16", false]], "w8a16_gptq (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A16_GPTQ", false]], "w8a8_sq_per_channel (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL", false]], "w8a8_sq_per_channel_per_tensor_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL_PER_TENSOR_PLUGIN", false]], "w8a8_sq_per_channel_per_token_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL_PER_TOKEN_PLUGIN", false]], "w8a8_sq_per_tensor_per_token_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_TENSOR_PER_TOKEN_PLUGIN", false]], "w8a8_sq_per_tensor_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[65, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_TENSOR_PLUGIN", false]], "weight_loader() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[78, "tensorrt_llm.layers.attention.DeepseekV2Attention.weight_loader", false]], "weight_loader() (tensorrt_llm.layers.embedding.embedding method)": [[78, "tensorrt_llm.layers.embedding.Embedding.weight_loader", false]], "weight_loader() (tensorrt_llm.layers.linear.linearbase method)": [[78, "tensorrt_llm.layers.linear.LinearBase.weight_loader", false]], "weight_sparsity (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.weight_sparsity", false]], "weight_streaming (tensorrt_llm.llmapi.buildconfig attribute)": [[65, "tensorrt_llm.llmapi.BuildConfig.weight_streaming", false]], "where() (in module tensorrt_llm.functional)": [[77, "tensorrt_llm.functional.where", false]], "whisperencoder (class in tensorrt_llm.models)": [[79, "tensorrt_llm.models.WhisperEncoder", false]], "workspace (tensorrt_llm.llmapi.llm attribute)": [[65, "tensorrt_llm.llmapi.LLM.workspace", false]], "workspace (tensorrt_llm.llmapi.llm property)": [[65, "id1", false]], "yarn (tensorrt_llm.functional.positionembeddingtype attribute)": [[77, "tensorrt_llm.functional.PositionEmbeddingType.yarn", false]], "yarn (tensorrt_llm.functional.rotaryscalingtype attribute)": [[77, "tensorrt_llm.functional.RotaryScalingType.yarn", false]]}, "objects": {"": [[1, 0, 1, "c.FMT_DIM", "FMT_DIM"], [1, 0, 1, "c.SET_FROM_OPTIONAL", "SET_FROM_OPTIONAL"], [1, 1, 1, "_CPPv48nvinfer1", "nvinfer1"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [0, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [0, 1, 1, "_CPPv4N12tensorrt_llm13batch_manager16kv_cache_managerE", "tensorrt_llm::batch_manager::kv_cache_manager"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutputE", "tensorrt_llm::executor::AdditionalModelOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput::gatherContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput::name"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput13gatherContextE", "tensorrt_llm::executor::AdditionalModelOutput::gatherContext"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput4nameE", "tensorrt_llm::executor::AdditionalModelOutput::name"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", "tensorrt_llm::executor::AdditionalModelOutput::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", "tensorrt_llm::executor::AdditionalModelOutput::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputE", "tensorrt_llm::executor::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::name"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::output"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput4nameE", "tensorrt_llm::executor::AdditionalOutput::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator=::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput6outputE", "tensorrt_llm::executor::AdditionalOutput::output"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputD0Ev", "tensorrt_llm::executor::AdditionalOutput::~AdditionalOutput"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12BatchingTypeE", "tensorrt_llm::executor::BatchingType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12BatchingType9kINFLIGHTE", "tensorrt_llm::executor::BatchingType::kINFLIGHT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12BatchingType7kSTATICE", "tensorrt_llm::executor::BatchingType::kSTATIC"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10BeamTokensE", "tensorrt_llm::executor::BeamTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10BufferViewE", "tensorrt_llm::executor::BufferView"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfigE", "tensorrt_llm::executor::CacheTransceiverConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", "tensorrt_llm::executor::CacheTransceiverConfig::CacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", "tensorrt_llm::executor::CacheTransceiverConfig::CacheTransceiverConfig::maxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfig15getMaxNumTokensEv", "tensorrt_llm::executor::CacheTransceiverConfig::getMaxNumTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig13mMaxNumTokensE", "tensorrt_llm::executor::CacheTransceiverConfig::mMaxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", "tensorrt_llm::executor::CacheTransceiverConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", "tensorrt_llm::executor::CacheTransceiverConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", "tensorrt_llm::executor::CacheTransceiverConfig::setMaxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", "tensorrt_llm::executor::CacheTransceiverConfig::setMaxNumTokens::maxNumTokens"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicyE", "tensorrt_llm::executor::CapacitySchedulerPolicy"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy20kGUARANTEED_NO_EVICTE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kGUARANTEED_NO_EVICT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy16kMAX_UTILIZATIONE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kMAX_UTILIZATION"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy13kSTATIC_BATCHE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kSTATIC_BATCH"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationModeE", "tensorrt_llm::executor::CommunicationMode"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationMode7kLEADERE", "tensorrt_llm::executor::CommunicationMode::kLEADER"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationMode13kORCHESTRATORE", "tensorrt_llm::executor::CommunicationMode::kORCHESTRATOR"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationTypeE", "tensorrt_llm::executor::CommunicationType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationType4kMPIE", "tensorrt_llm::executor::CommunicationType::kMPI"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicyE", "tensorrt_llm::executor::ContextChunkingPolicy"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy15kEQUAL_PROGRESSE", "tensorrt_llm::executor::ContextChunkingPolicy::kEQUAL_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy24kFIRST_COME_FIRST_SERVEDE", "tensorrt_llm::executor::ContextChunkingPolicy::kFIRST_COME_FIRST_SERVED"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsE", "tensorrt_llm::executor::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERR18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::serializedState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::state"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams13RequestIdTypeE", "tensorrt_llm::executor::ContextPhaseParams::RequestIdType"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8StatePtrE", "tensorrt_llm::executor::ContextPhaseParams::StatePtr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", "tensorrt_llm::executor::ContextPhaseParams::deleter"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", "tensorrt_llm::executor::ContextPhaseParams::deleter::data"], [0, 3, 1, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams14getDraftTokensEv", "tensorrt_llm::executor::ContextPhaseParams::getDraftTokens"], [0, 3, 1, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams17getFirstGenTokensEv", "tensorrt_llm::executor::ContextPhaseParams::getFirstGenTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getReqIdEv", "tensorrt_llm::executor::ContextPhaseParams::getReqId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams18getSerializedStateEv", "tensorrt_llm::executor::ContextPhaseParams::getSerializedState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8getStateEv", "tensorrt_llm::executor::ContextPhaseParams::getState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getStateEv", "tensorrt_llm::executor::ContextPhaseParams::getState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12mDraftTokensE", "tensorrt_llm::executor::ContextPhaseParams::mDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams15mFirstGenTokensE", "tensorrt_llm::executor::ContextPhaseParams::mFirstGenTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mReqIdE", "tensorrt_llm::executor::ContextPhaseParams::mReqId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mStateE", "tensorrt_llm::executor::ContextPhaseParams::mState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERR18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator="], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParamseqERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator=="], [0, 3, 1, "_CPPv4NO12tensorrt_llm8executor18ContextPhaseParams17popFirstGenTokensEv", "tensorrt_llm::executor::ContextPhaseParams::popFirstGenTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12releaseStateEv", "tensorrt_llm::executor::ContextPhaseParams::releaseState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsD0Ev", "tensorrt_llm::executor::ContextPhaseParams::~ContextPhaseParams"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverStateE", "tensorrt_llm::executor::DataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEv", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState::cacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState::commState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState13getCacheStateEv", "tensorrt_llm::executor::DataTransceiverState::getCacheState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState12getCommStateEv", "tensorrt_llm::executor::DataTransceiverState::getCommState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState11mCacheStateE", "tensorrt_llm::executor::DataTransceiverState::mCacheState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState10mCommStateE", "tensorrt_llm::executor::DataTransceiverState::mCommState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", "tensorrt_llm::executor::DataTransceiverState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", "tensorrt_llm::executor::DataTransceiverState::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", "tensorrt_llm::executor::DataTransceiverState::setCacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", "tensorrt_llm::executor::DataTransceiverState::setCacheState::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::setCommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::setCommState::state"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState8toStringEv", "tensorrt_llm::executor::DataTransceiverState::toString"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8DataTypeE", "tensorrt_llm::executor::DataType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kBF16E", "tensorrt_llm::executor::DataType::kBF16"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kBOOLE", "tensorrt_llm::executor::DataType::kBOOL"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kFP16E", "tensorrt_llm::executor::DataType::kFP16"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kFP32E", "tensorrt_llm::executor::DataType::kFP32"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType4kFP8E", "tensorrt_llm::executor::DataType::kFP8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kINT32E", "tensorrt_llm::executor::DataType::kINT32"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kINT64E", "tensorrt_llm::executor::DataType::kINT64"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kINT8E", "tensorrt_llm::executor::DataType::kINT8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kUINT8E", "tensorrt_llm::executor::DataType::kUINT8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType8kUNKNOWNE", "tensorrt_llm::executor::DataType::kUNKNOWN"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfigE", "tensorrt_llm::executor::DebugConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugInputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugOutputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugTensorNames"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugTensorsMaxIterations"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig9StringVecE", "tensorrt_llm::executor::DebugConfig::StringVec"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig20getDebugInputTensorsEv", "tensorrt_llm::executor::DebugConfig::getDebugInputTensors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig21getDebugOutputTensorsEv", "tensorrt_llm::executor::DebugConfig::getDebugOutputTensors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig19getDebugTensorNamesEv", "tensorrt_llm::executor::DebugConfig::getDebugTensorNames"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig28getDebugTensorsMaxIterationsEv", "tensorrt_llm::executor::DebugConfig::getDebugTensorsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig18mDebugInputTensorsE", "tensorrt_llm::executor::DebugConfig::mDebugInputTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19mDebugOutputTensorsE", "tensorrt_llm::executor::DebugConfig::mDebugOutputTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig17mDebugTensorNamesE", "tensorrt_llm::executor::DebugConfig::mDebugTensorNames"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig26mDebugTensorsMaxIterationsE", "tensorrt_llm::executor::DebugConfig::mDebugTensorsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", "tensorrt_llm::executor::DebugConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", "tensorrt_llm::executor::DebugConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugInputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugInputTensors::debugInputTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugOutputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugOutputTensors::debugOutputTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", "tensorrt_llm::executor::DebugConfig::setDebugTensorNames"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", "tensorrt_llm::executor::DebugConfig::setDebugTensorNames::debugTensorNames"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", "tensorrt_llm::executor::DebugConfig::setDebugTensorsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", "tensorrt_llm::executor::DebugConfig::setDebugTensorsMaxIterations::debugTensorsMaxIterations"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIterationE", "tensorrt_llm::executor::DebugTensorsPerIteration"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration12debugTensorsE", "tensorrt_llm::executor::DebugTensorsPerIteration::debugTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration4iterE", "tensorrt_llm::executor::DebugTensorsPerIteration::iter"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfigE", "tensorrt_llm::executor::DecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::medusaChoices"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31enableSeamlessLookaheadDecodingEv", "tensorrt_llm::executor::DecodingConfig::enableSeamlessLookaheadDecoding"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig15getDecodingModeEv", "tensorrt_llm::executor::DecodingConfig::getDecodingMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig14getEagleConfigEv", "tensorrt_llm::executor::DecodingConfig::getEagleConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig26getLookaheadDecodingConfigEv", "tensorrt_llm::executor::DecodingConfig::getLookaheadDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig33getLookaheadDecodingMaxNumRequestEv", "tensorrt_llm::executor::DecodingConfig::getLookaheadDecodingMaxNumRequest"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig16getMedusaChoicesEv", "tensorrt_llm::executor::DecodingConfig::getMedusaChoices"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig13mDecodingModeE", "tensorrt_llm::executor::DecodingConfig::mDecodingMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig12mEagleConfigE", "tensorrt_llm::executor::DecodingConfig::mEagleConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig24mLookaheadDecodingConfigE", "tensorrt_llm::executor::DecodingConfig::mLookaheadDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31mLookaheadDecodingMaxNumRequestE", "tensorrt_llm::executor::DecodingConfig::mLookaheadDecodingMaxNumRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14mMedusaChoicesE", "tensorrt_llm::executor::DecodingConfig::mMedusaChoices"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", "tensorrt_llm::executor::DecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", "tensorrt_llm::executor::DecodingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig15setDecodingModeERK12DecodingMode", "tensorrt_llm::executor::DecodingConfig::setDecodingMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14setEagleConfigERK11EagleConfig", "tensorrt_llm::executor::DecodingConfig::setEagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::DecodingConfig::setLookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::DecodingConfig::setLookaheadDecodingConfig::lookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig16setMedusaChoicesERK13MedusaChoices", "tensorrt_llm::executor::DecodingConfig::setMedusaChoices"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12DecodingModeE", "tensorrt_llm::executor::DecodingMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4AutoEv", "tensorrt_llm::executor::DecodingMode::Auto"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode10BeamSearchEv", "tensorrt_llm::executor::DecodingMode::BeamSearch"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::DecodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::DecodingMode::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5EagleEv", "tensorrt_llm::executor::DecodingMode::Eagle"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExplicitDraftTokensEv", "tensorrt_llm::executor::DecodingMode::ExplicitDraftTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExternalDraftTokensEv", "tensorrt_llm::executor::DecodingMode::ExternalDraftTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9LookaheadEv", "tensorrt_llm::executor::DecodingMode::Lookahead"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6MedusaEv", "tensorrt_llm::executor::DecodingMode::Medusa"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopKEv", "tensorrt_llm::executor::DecodingMode::TopK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8TopKTopPEv", "tensorrt_llm::executor::DecodingMode::TopKTopP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopPEv", "tensorrt_llm::executor::DecodingMode::TopP"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14UnderlyingTypeE", "tensorrt_llm::executor::DecodingMode::UnderlyingType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::allBitSet"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::allBitSet::bits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::anyBitSet"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::anyBitSet::bits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7getNameEv", "tensorrt_llm::executor::DecodingMode::getName"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8getStateEv", "tensorrt_llm::executor::DecodingMode::getState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isAutoEv", "tensorrt_llm::executor::DecodingMode::isAuto"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isBeamSearchEv", "tensorrt_llm::executor::DecodingMode::isBeamSearch"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7isEagleEv", "tensorrt_llm::executor::DecodingMode::isEagle"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExplicitDraftTokensEv", "tensorrt_llm::executor::DecodingMode::isExplicitDraftTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExternalDraftTokensEv", "tensorrt_llm::executor::DecodingMode::isExternalDraftTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode11isLookaheadEv", "tensorrt_llm::executor::DecodingMode::isLookahead"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8isMedusaEv", "tensorrt_llm::executor::DecodingMode::isMedusa"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopKEv", "tensorrt_llm::executor::DecodingMode::isTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isTopKandTopPEv", "tensorrt_llm::executor::DecodingMode::isTopKandTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isTopKorTopPEv", "tensorrt_llm::executor::DecodingMode::isTopKorTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopPEv", "tensorrt_llm::executor::DecodingMode::isTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseBanTokensEv", "tensorrt_llm::executor::DecodingMode::isUseBanTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isUseBanWordsEv", "tensorrt_llm::executor::DecodingMode::isUseBanWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUseExplicitEosStopEv", "tensorrt_llm::executor::DecodingMode::isUseExplicitEosStop"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isUseFrequencyPenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseFrequencyPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode18isUseMaxLengthStopEv", "tensorrt_llm::executor::DecodingMode::isUseMaxLengthStop"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseMinLengthEv", "tensorrt_llm::executor::DecodingMode::isUseMinLength"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9isUseMinPEv", "tensorrt_llm::executor::DecodingMode::isUseMinP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseNoRepeatNgramSizeEv", "tensorrt_llm::executor::DecodingMode::isUseNoRepeatNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseOccurrencePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseOccurrencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isUsePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUsePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUsePresencePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUsePresencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseRepetitionPenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseRepetitionPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode17isUseStopCriteriaEv", "tensorrt_llm::executor::DecodingMode::isUseStopCriteria"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseStopWordsEv", "tensorrt_llm::executor::DecodingMode::isUseStopWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode16isUseTemperatureEv", "tensorrt_llm::executor::DecodingMode::isUseTemperature"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode28isUseVariableBeamWidthSearchEv", "tensorrt_llm::executor::DecodingMode::isUseVariableBeamWidthSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kAutoE", "tensorrt_llm::executor::DecodingMode::kAuto"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11kBeamSearchE", "tensorrt_llm::executor::DecodingMode::kBeamSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6kEagleE", "tensorrt_llm::executor::DecodingMode::kEagle"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExplicitDraftTokensE", "tensorrt_llm::executor::DecodingMode::kExplicitDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExternalDraftTokensE", "tensorrt_llm::executor::DecodingMode::kExternalDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode10kLookaheadE", "tensorrt_llm::executor::DecodingMode::kLookahead"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7kMedusaE", "tensorrt_llm::executor::DecodingMode::kMedusa"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kNumFlagsE", "tensorrt_llm::executor::DecodingMode::kNumFlags"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopKE", "tensorrt_llm::executor::DecodingMode::kTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kTopKTopPE", "tensorrt_llm::executor::DecodingMode::kTopKTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopPE", "tensorrt_llm::executor::DecodingMode::kTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseBanTokensE", "tensorrt_llm::executor::DecodingMode::kUseBanTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12kUseBanWordsE", "tensorrt_llm::executor::DecodingMode::kUseBanWords"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19kUseExplicitEosStopE", "tensorrt_llm::executor::DecodingMode::kUseExplicitEosStop"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22kUseFrequencyPenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseFrequencyPenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode17kUseMaxLengthStopE", "tensorrt_llm::executor::DecodingMode::kUseMaxLengthStop"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseMinLengthE", "tensorrt_llm::executor::DecodingMode::kUseMinLength"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8kUseMinPE", "tensorrt_llm::executor::DecodingMode::kUseMinP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUseNoRepeatNgramSizeE", "tensorrt_llm::executor::DecodingMode::kUseNoRepeatNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseOccurrencePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseOccurrencePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUsePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUsePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUsePresencePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUsePresencePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseRepetitionPenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseRepetitionPenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode24kUseStandardStopCriteriaE", "tensorrt_llm::executor::DecodingMode::kUseStandardStopCriteria"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseStopWordsE", "tensorrt_llm::executor::DecodingMode::kUseStopWords"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode15kUseTemperatureE", "tensorrt_llm::executor::DecodingMode::kUseTemperature"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode27kUseVariableBeamWidthSearchE", "tensorrt_llm::executor::DecodingMode::kUseVariableBeamWidthSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6mStateE", "tensorrt_llm::executor::DecodingMode::mState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", "tensorrt_llm::executor::DecodingMode::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", "tensorrt_llm::executor::DecodingMode::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo::x"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", "tensorrt_llm::executor::DecodingMode::useBanTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", "tensorrt_llm::executor::DecodingMode::useBanTokens::banTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", "tensorrt_llm::executor::DecodingMode::useBanWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", "tensorrt_llm::executor::DecodingMode::useBanWords::banWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", "tensorrt_llm::executor::DecodingMode::useExplicitEosStop"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", "tensorrt_llm::executor::DecodingMode::useExplicitEosStop::explicitEosStop"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", "tensorrt_llm::executor::DecodingMode::useFrequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", "tensorrt_llm::executor::DecodingMode::useFrequencyPenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", "tensorrt_llm::executor::DecodingMode::useMaxLengthStop"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", "tensorrt_llm::executor::DecodingMode::useMaxLengthStop::maxLengthStop"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", "tensorrt_llm::executor::DecodingMode::useMinLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", "tensorrt_llm::executor::DecodingMode::useMinLength::useMinLen"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", "tensorrt_llm::executor::DecodingMode::useMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", "tensorrt_llm::executor::DecodingMode::useMinP::useMinP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", "tensorrt_llm::executor::DecodingMode::useNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", "tensorrt_llm::executor::DecodingMode::useNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", "tensorrt_llm::executor::DecodingMode::useOccurrencePenalties"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", "tensorrt_llm::executor::DecodingMode::useOccurrencePenalties::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", "tensorrt_llm::executor::DecodingMode::usePresencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", "tensorrt_llm::executor::DecodingMode::usePresencePenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", "tensorrt_llm::executor::DecodingMode::useRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", "tensorrt_llm::executor::DecodingMode::useRepetitionPenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", "tensorrt_llm::executor::DecodingMode::useStopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", "tensorrt_llm::executor::DecodingMode::useStopWords::stopWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", "tensorrt_llm::executor::DecodingMode::useTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", "tensorrt_llm::executor::DecodingMode::useTemperature::useTemp"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", "tensorrt_llm::executor::DecodingMode::useVariableBeamWidthSearch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", "tensorrt_llm::executor::DecodingMode::useVariableBeamWidthSearch::useVariableBeamWidthSearch"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStatsE", "tensorrt_llm::executor::DisServingRequestStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats11kvCacheSizeE", "tensorrt_llm::executor::DisServingRequestStats::kvCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats17kvCacheTransferMSE", "tensorrt_llm::executor::DisServingRequestStats::kvCacheTransferMS"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfigE", "tensorrt_llm::executor::DynamicBatchConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::batchSizeTable"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::dynamicBatchMovingAverageWindow"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::enableBatchSizeTuning"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::enableMaxNumTokensTuning"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig17getBatchSizeTableEv", "tensorrt_llm::executor::DynamicBatchConfig::getBatchSizeTable"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig34getDynamicBatchMovingAverageWindowEv", "tensorrt_llm::executor::DynamicBatchConfig::getDynamicBatchMovingAverageWindow"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig24getEnableBatchSizeTuningEv", "tensorrt_llm::executor::DynamicBatchConfig::getEnableBatchSizeTuning"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig27getEnableMaxNumTokensTuningEv", "tensorrt_llm::executor::DynamicBatchConfig::getEnableMaxNumTokensTuning"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22kDefaultBatchSizeTableE", "tensorrt_llm::executor::DynamicBatchConfig::kDefaultBatchSizeTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig39kDefaultDynamicBatchMovingAverageWindowE", "tensorrt_llm::executor::DynamicBatchConfig::kDefaultDynamicBatchMovingAverageWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig15mBatchSizeTableE", "tensorrt_llm::executor::DynamicBatchConfig::mBatchSizeTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig32mDynamicBatchMovingAverageWindowE", "tensorrt_llm::executor::DynamicBatchConfig::mDynamicBatchMovingAverageWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22mEnableBatchSizeTuningE", "tensorrt_llm::executor::DynamicBatchConfig::mEnableBatchSizeTuning"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig25mEnableMaxNumTokensTuningE", "tensorrt_llm::executor::DynamicBatchConfig::mEnableMaxNumTokensTuning"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12EagleChoicesE", "tensorrt_llm::executor::EagleChoices"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfigE", "tensorrt_llm::executor::EagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::dynamicTreeMaxTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::eagleChoices"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::greedySampling"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::posteriorThreshold"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::useDynamicTree"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", "tensorrt_llm::executor::EagleConfig::checkPosteriorValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", "tensorrt_llm::executor::EagleConfig::checkPosteriorValue::value"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getDynamicTreeMaxTopKEv", "tensorrt_llm::executor::EagleConfig::getDynamicTreeMaxTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig15getEagleChoicesEv", "tensorrt_llm::executor::EagleConfig::getEagleChoices"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getPosteriorThresholdEv", "tensorrt_llm::executor::EagleConfig::getPosteriorThreshold"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig16isGreedySamplingEv", "tensorrt_llm::executor::EagleConfig::isGreedySampling"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mDynamicTreeMaxTopKE", "tensorrt_llm::executor::EagleConfig::mDynamicTreeMaxTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig13mEagleChoicesE", "tensorrt_llm::executor::EagleConfig::mEagleChoices"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mGreedySamplingE", "tensorrt_llm::executor::EagleConfig::mGreedySampling"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mPosteriorThresholdE", "tensorrt_llm::executor::EagleConfig::mPosteriorThreshold"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mUseDynamicTreeE", "tensorrt_llm::executor::EagleConfig::mUseDynamicTree"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", "tensorrt_llm::executor::EagleConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", "tensorrt_llm::executor::EagleConfig::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig14useDynamicTreeEv", "tensorrt_llm::executor::EagleConfig::useDynamicTree"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8ExecutorE", "tensorrt_llm::executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERR8Executor", "tensorrt_llm::executor::Executor::Executor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderEngineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderJsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderModel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderModelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderEngineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderJsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderModel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderModelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::engineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", "tensorrt_llm::executor::Executor::Executor::executor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::jsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::managedWeights"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::model"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::requestId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::requestIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor18canEnqueueRequestsEv", "tensorrt_llm::executor::Executor::canEnqueueRequests"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", "tensorrt_llm::executor::Executor::cancelRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", "tensorrt_llm::executor::Executor::cancelRequest::requestId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", "tensorrt_llm::executor::Executor::enqueueRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", "tensorrt_llm::executor::Executor::enqueueRequest::request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", "tensorrt_llm::executor::Executor::enqueueRequests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", "tensorrt_llm::executor::Executor::enqueueRequests::requests"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor22getKVCacheEventManagerEv", "tensorrt_llm::executor::Executor::getKVCacheEventManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestDebugTensorsEv", "tensorrt_llm::executor::Executor::getLatestDebugTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor23getLatestIterationStatsEv", "tensorrt_llm::executor::Executor::getLatestIterationStats"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestRequestStatsEv", "tensorrt_llm::executor::Executor::getLatestRequestStats"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Executor::getNumResponsesReady"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Executor::getNumResponsesReady::requestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor13isParticipantEv", "tensorrt_llm::executor::Executor::isParticipant"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8Executor5mImplE", "tensorrt_llm::executor::Executor::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", "tensorrt_llm::executor::Executor::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERR8Executor", "tensorrt_llm::executor::Executor::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", "tensorrt_llm::executor::Executor::operator=::executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8shutdownEv", "tensorrt_llm::executor::Executor::shutdown"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutorD0Ev", "tensorrt_llm::executor::Executor::~Executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfigE", "tensorrt_llm::executor::ExecutorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::additionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::batchingType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::enableChunkedContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::enableTrtOverlap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::gatherGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::gpuWeightsPercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::iterStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::logitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxBatchSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxQueueSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxSeqIdleMicroseconds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::normalizeLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::promptTableOffloading"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::recvPollPeriodMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::requestStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::useGpuDirectStorage"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getAdditionalModelOutputsEv", "tensorrt_llm::executor::ExecutorConfig::getAdditionalModelOutputs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getBatchingTypeEv", "tensorrt_llm::executor::ExecutorConfig::getBatchingType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getCacheTransceiverConfigEv", "tensorrt_llm::executor::ExecutorConfig::getCacheTransceiverConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig14getDebugConfigEv", "tensorrt_llm::executor::ExecutorConfig::getDebugConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getDecodingConfigEv", "tensorrt_llm::executor::ExecutorConfig::getDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getEnableChunkedContextEv", "tensorrt_llm::executor::ExecutorConfig::getEnableChunkedContext"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getEnableTrtOverlapEv", "tensorrt_llm::executor::ExecutorConfig::getEnableTrtOverlap"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig32getExtendedRuntimePerfKnobConfigEv", "tensorrt_llm::executor::ExecutorConfig::getExtendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getGatherGenerationLogitsEv", "tensorrt_llm::executor::ExecutorConfig::getGatherGenerationLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getGpuWeightsPercentEv", "tensorrt_llm::executor::ExecutorConfig::getGpuWeightsPercent"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getGuidedDecodingConfigEv", "tensorrt_llm::executor::ExecutorConfig::getGuidedDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getIterStatsMaxIterationsEv", "tensorrt_llm::executor::ExecutorConfig::getIterStatsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getKvCacheConfigEv", "tensorrt_llm::executor::ExecutorConfig::getKvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19getKvCacheConfigRefEv", "tensorrt_llm::executor::ExecutorConfig::getKvCacheConfigRef"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getLogitsPostProcessorConfigEv", "tensorrt_llm::executor::ExecutorConfig::getLogitsPostProcessorConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBatchSizeEv", "tensorrt_llm::executor::ExecutorConfig::getMaxBatchSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBeamWidthEv", "tensorrt_llm::executor::ExecutorConfig::getMaxBeamWidth"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxNumTokensEv", "tensorrt_llm::executor::ExecutorConfig::getMaxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxQueueSizeEv", "tensorrt_llm::executor::ExecutorConfig::getMaxQueueSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getMaxSeqIdleMicrosecondsEv", "tensorrt_llm::executor::ExecutorConfig::getMaxSeqIdleMicroseconds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getNormalizeLogProbsEv", "tensorrt_llm::executor::ExecutorConfig::getNormalizeLogProbs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getParallelConfigEv", "tensorrt_llm::executor::ExecutorConfig::getParallelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getPeftCacheConfigEv", "tensorrt_llm::executor::ExecutorConfig::getPeftCacheConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig24getPromptTableOffloadingEv", "tensorrt_llm::executor::ExecutorConfig::getPromptTableOffloading"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getRecvPollPeriodMsEv", "tensorrt_llm::executor::ExecutorConfig::getRecvPollPeriodMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getRequestStatsMaxIterationsEv", "tensorrt_llm::executor::ExecutorConfig::getRequestStatsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getSchedulerConfigEv", "tensorrt_llm::executor::ExecutorConfig::getSchedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21getSchedulerConfigRefEv", "tensorrt_llm::executor::ExecutorConfig::getSchedulerConfigRef"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getSpecDecConfigEv", "tensorrt_llm::executor::ExecutorConfig::getSpecDecConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig22getUseGpuDirectStorageEv", "tensorrt_llm::executor::ExecutorConfig::getUseGpuDirectStorage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultIterStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultIterStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultMaxSeqIdleMicrosecondsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultMaxSeqIdleMicroseconds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig33kDefaultRequestStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultRequestStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mAdditionalModelOutputsE", "tensorrt_llm::executor::ExecutorConfig::mAdditionalModelOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mBatchingTypeE", "tensorrt_llm::executor::ExecutorConfig::mBatchingType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mCacheTransceiverConfigE", "tensorrt_llm::executor::ExecutorConfig::mCacheTransceiverConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig12mDebugConfigE", "tensorrt_llm::executor::ExecutorConfig::mDebugConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mEnableChunkedContextE", "tensorrt_llm::executor::ExecutorConfig::mEnableChunkedContext"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mEnableTrtOverlapE", "tensorrt_llm::executor::ExecutorConfig::mEnableTrtOverlap"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30mExtendedRuntimePerfKnobConfigE", "tensorrt_llm::executor::ExecutorConfig::mExtendedRuntimePerfKnobConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mGatherGenerationLogitsE", "tensorrt_llm::executor::ExecutorConfig::mGatherGenerationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mGpuWeightsPercentE", "tensorrt_llm::executor::ExecutorConfig::mGpuWeightsPercent"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mGuidedDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mGuidedDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mIterStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::mIterStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14mKvCacheConfigE", "tensorrt_llm::executor::ExecutorConfig::mKvCacheConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mLogitsPostProcessorConfigE", "tensorrt_llm::executor::ExecutorConfig::mLogitsPostProcessorConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBatchSizeE", "tensorrt_llm::executor::ExecutorConfig::mMaxBatchSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBeamWidthE", "tensorrt_llm::executor::ExecutorConfig::mMaxBeamWidth"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxNumTokensE", "tensorrt_llm::executor::ExecutorConfig::mMaxNumTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxQueueSizeE", "tensorrt_llm::executor::ExecutorConfig::mMaxQueueSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mMaxSeqIdleMicrosecondsE", "tensorrt_llm::executor::ExecutorConfig::mMaxSeqIdleMicroseconds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mNormalizeLogProbsE", "tensorrt_llm::executor::ExecutorConfig::mNormalizeLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mParallelConfigE", "tensorrt_llm::executor::ExecutorConfig::mParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mPeftCacheConfigE", "tensorrt_llm::executor::ExecutorConfig::mPeftCacheConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22mPromptTableOffloadingE", "tensorrt_llm::executor::ExecutorConfig::mPromptTableOffloading"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mRecvPollPeriodMsE", "tensorrt_llm::executor::ExecutorConfig::mRecvPollPeriodMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mRequestStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::mRequestStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mSchedulerConfigE", "tensorrt_llm::executor::ExecutorConfig::mSchedulerConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mSpeculativeDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mSpeculativeDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20mUseGpuDirectStorageE", "tensorrt_llm::executor::ExecutorConfig::mUseGpuDirectStorage"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", "tensorrt_llm::executor::ExecutorConfig::setAdditionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", "tensorrt_llm::executor::ExecutorConfig::setAdditionalModelOutputs::additionalModelOutputs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", "tensorrt_llm::executor::ExecutorConfig::setBatchingType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", "tensorrt_llm::executor::ExecutorConfig::setBatchingType::batchingType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", "tensorrt_llm::executor::ExecutorConfig::setCacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", "tensorrt_llm::executor::ExecutorConfig::setCacheTransceiverConfig::cacheTransceiverConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", "tensorrt_llm::executor::ExecutorConfig::setDebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", "tensorrt_llm::executor::ExecutorConfig::setDebugConfig::debugConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setDecodingConfig::decodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", "tensorrt_llm::executor::ExecutorConfig::setEnableChunkedContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", "tensorrt_llm::executor::ExecutorConfig::setEnableChunkedContext::enableChunkedContext"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", "tensorrt_llm::executor::ExecutorConfig::setEnableTrtOverlap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", "tensorrt_llm::executor::ExecutorConfig::setEnableTrtOverlap::enableTrtOverlap"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExecutorConfig::setExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExecutorConfig::setExtendedRuntimePerfKnobConfig::extendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", "tensorrt_llm::executor::ExecutorConfig::setGatherGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", "tensorrt_llm::executor::ExecutorConfig::setGatherGenerationLogits::gatherGenerationLogits"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", "tensorrt_llm::executor::ExecutorConfig::setGpuWeightsPercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", "tensorrt_llm::executor::ExecutorConfig::setGpuWeightsPercent::gpuWeightsPercent"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setGuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setGuidedDecodingConfig::guidedDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setIterStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setIterStatsMaxIterations::iterStatsMaxIterations"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setKvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setKvCacheConfig::kvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", "tensorrt_llm::executor::ExecutorConfig::setLogitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", "tensorrt_llm::executor::ExecutorConfig::setLogitsPostProcessorConfig::logitsPostProcessorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBatchSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBatchSize::maxBatchSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBeamWidth::maxBeamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxNumTokens::maxNumTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ExecutorConfig::setMaxQueueSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ExecutorConfig::setMaxQueueSize::maxQueueSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", "tensorrt_llm::executor::ExecutorConfig::setMaxSeqIdleMicroseconds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", "tensorrt_llm::executor::ExecutorConfig::setMaxSeqIdleMicroseconds::maxSeqIdleMicroseconds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", "tensorrt_llm::executor::ExecutorConfig::setNormalizeLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", "tensorrt_llm::executor::ExecutorConfig::setNormalizeLogProbs::normalizeLogProbs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", "tensorrt_llm::executor::ExecutorConfig::setParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", "tensorrt_llm::executor::ExecutorConfig::setParallelConfig::parallelConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setPeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setPeftCacheConfig::peftCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", "tensorrt_llm::executor::ExecutorConfig::setPromptTableOffloading"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", "tensorrt_llm::executor::ExecutorConfig::setPromptTableOffloading::promptTableOffloading"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRecvPollPeriodMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRecvPollPeriodMs::recvPollPeriodMs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRequestStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRequestStatsMaxIterations::requestStatsMaxIterations"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", "tensorrt_llm::executor::ExecutorConfig::setSchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", "tensorrt_llm::executor::ExecutorConfig::setSchedulerConfig::schedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setSpecDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setSpecDecConfig::specDecConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", "tensorrt_llm::executor::ExecutorConfig::setUseGpuDirectStorage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", "tensorrt_llm::executor::ExecutorConfig::setUseGpuDirectStorage::useGpuDirectStorage"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::cudaGraphCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::cudaGraphMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::enableContextFMHAFP32Acc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::multiBlockMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21getCudaGraphCacheSizeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getCudaGraphCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16getCudaGraphModeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getCudaGraphMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27getEnableContextFMHAFP32AccEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getEnableContextFMHAFP32Acc"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17getMultiBlockModeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getMultiBlockMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig19mCudaGraphCacheSizeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mCudaGraphCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig14mCudaGraphModeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mCudaGraphMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig25mEnableContextFMHAFP32AccE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mEnableContextFMHAFP32Acc"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig15mMultiBlockModeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mMultiBlockMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphCacheSize::cacheSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphMode::cudaGraphMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setEnableContextFMHAFP32Acc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setEnableContextFMHAFP32Acc::enableContextFMHAFP32Acc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setMultiBlockMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setMultiBlockMode::multiBlockMode"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfigE", "tensorrt_llm::executor::ExternalDraftTokensConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::acceptanceThreshold"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::fastLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::logits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::tokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig22getAcceptanceThresholdEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getAcceptanceThreshold"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig13getFastLogitsEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getFastLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getLogitsEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getTokensEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig20mAcceptanceThresholdE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mAcceptanceThreshold"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig11mFastLogitsE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mFastLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mLogitsE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mTokensE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mTokens"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12FinishReasonE", "tensorrt_llm::executor::FinishReason"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason10kCANCELLEDE", "tensorrt_llm::executor::FinishReason::kCANCELLED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason7kEND_IDE", "tensorrt_llm::executor::FinishReason::kEND_ID"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason7kLENGTHE", "tensorrt_llm::executor::FinishReason::kLENGTH"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason13kNOT_FINISHEDE", "tensorrt_llm::executor::FinishReason::kNOT_FINISHED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason11kSTOP_WORDSE", "tensorrt_llm::executor::FinishReason::kSTOP_WORDS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason10kTIMED_OUTE", "tensorrt_llm::executor::FinishReason::kTIMED_OUT"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9FloatTypeE", "tensorrt_llm::executor::FloatType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfigE", "tensorrt_llm::executor::GuidedDecodingConfig"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackendE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingBackend"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackend9kXGRAMMARE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingBackend::kXGRAMMAR"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::backend"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::encodedVocab"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::stopTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::tokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig10getBackendEv", "tensorrt_llm::executor::GuidedDecodingConfig::getBackend"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getEncodedVocabEv", "tensorrt_llm::executor::GuidedDecodingConfig::getEncodedVocab"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getStopTokenIdsEv", "tensorrt_llm::executor::GuidedDecodingConfig::getStopTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getTokenizerStrEv", "tensorrt_llm::executor::GuidedDecodingConfig::getTokenizerStr"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig8mBackendE", "tensorrt_llm::executor::GuidedDecodingConfig::mBackend"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mEncodedVocabE", "tensorrt_llm::executor::GuidedDecodingConfig::mEncodedVocab"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mStopTokenIdsE", "tensorrt_llm::executor::GuidedDecodingConfig::mStopTokenIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mTokenizerStrE", "tensorrt_llm::executor::GuidedDecodingConfig::mTokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", "tensorrt_llm::executor::GuidedDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", "tensorrt_llm::executor::GuidedDecodingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", "tensorrt_llm::executor::GuidedDecodingConfig::setBackend"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", "tensorrt_llm::executor::GuidedDecodingConfig::setBackend::backend"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingConfig::setEncodedVocab"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingConfig::setEncodedVocab::encodedVocab"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", "tensorrt_llm::executor::GuidedDecodingConfig::setStopTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", "tensorrt_llm::executor::GuidedDecodingConfig::setStopTokenIds::stopTokenIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", "tensorrt_llm::executor::GuidedDecodingConfig::setTokenizerStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", "tensorrt_llm::executor::GuidedDecodingConfig::setTokenizerStr::tokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig8validateEv", "tensorrt_llm::executor::GuidedDecodingConfig::validate"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParamsE", "tensorrt_llm::executor::GuidedDecodingParams"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideTypeE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType13kEBNF_GRAMMARE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kEBNF_GRAMMAR"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType5kJSONE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kJSON"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType12kJSON_SCHEMAE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kJSON_SCHEMA"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType6kREGEXE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kREGEX"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType15kSTRUCTURAL_TAGE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kSTRUCTURAL_TAG"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams::guide"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams::guideType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams8getGuideEv", "tensorrt_llm::executor::GuidedDecodingParams::getGuide"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams12getGuideTypeEv", "tensorrt_llm::executor::GuidedDecodingParams::getGuideType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams6mGuideE", "tensorrt_llm::executor::GuidedDecodingParams::mGuide"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams10mGuideTypeE", "tensorrt_llm::executor::GuidedDecodingParams::mGuideType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", "tensorrt_llm::executor::GuidedDecodingParams::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", "tensorrt_llm::executor::GuidedDecodingParams::operator==::other"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6IdTypeE", "tensorrt_llm::executor::IdType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStatsE", "tensorrt_llm::executor::InflightBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats26avgNumDecodedTokensPerIterE", "tensorrt_llm::executor::InflightBatchingStats::avgNumDecodedTokensPerIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12microBatchIdE", "tensorrt_llm::executor::InflightBatchingStats::microBatchId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats18numContextRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numContextRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12numCtxTokensE", "tensorrt_llm::executor::InflightBatchingStats::numCtxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats14numGenRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numGenRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats17numPausedRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numPausedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats20numScheduledRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numScheduledRequests"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14IterationStatsE", "tensorrt_llm::executor::IterationStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats11cpuMemUsageE", "tensorrt_llm::executor::IterationStats::cpuMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17crossKvCacheStatsE", "tensorrt_llm::executor::IterationStats::crossKvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats11gpuMemUsageE", "tensorrt_llm::executor::IterationStats::gpuMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats21inflightBatchingStatsE", "tensorrt_llm::executor::IterationStats::inflightBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats4iterE", "tensorrt_llm::executor::IterationStats::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats13iterLatencyMSE", "tensorrt_llm::executor::IterationStats::iterLatencyMS"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats12kvCacheStatsE", "tensorrt_llm::executor::IterationStats::kvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxBatchSizeRuntimeE", "tensorrt_llm::executor::IterationStats::maxBatchSizeRuntime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxBatchSizeStaticE", "tensorrt_llm::executor::IterationStats::maxBatchSizeStatic"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxBatchSizeTunerRecommendedE", "tensorrt_llm::executor::IterationStats::maxBatchSizeTunerRecommended"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20maxNumActiveRequestsE", "tensorrt_llm::executor::IterationStats::maxNumActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxNumTokensRuntimeE", "tensorrt_llm::executor::IterationStats::maxNumTokensRuntime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxNumTokensStaticE", "tensorrt_llm::executor::IterationStats::maxNumTokensStatic"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxNumTokensTunerRecommendedE", "tensorrt_llm::executor::IterationStats::maxNumTokensTunerRecommended"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats31newActiveRequestsQueueLatencyMSE", "tensorrt_llm::executor::IterationStats::newActiveRequestsQueueLatencyMS"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17numActiveRequestsE", "tensorrt_llm::executor::IterationStats::numActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20numCompletedRequestsE", "tensorrt_llm::executor::IterationStats::numCompletedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20numNewActiveRequestsE", "tensorrt_llm::executor::IterationStats::numNewActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17numQueuedRequestsE", "tensorrt_llm::executor::IterationStats::numQueuedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats14pinnedMemUsageE", "tensorrt_llm::executor::IterationStats::pinnedMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19staticBatchingStatsE", "tensorrt_llm::executor::IterationStats::staticBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats9timestampE", "tensorrt_llm::executor::IterationStats::timestamp"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor13IterationTypeE", "tensorrt_llm::executor::IterationType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerializationE", "tensorrt_llm::executor::JsonSerialization"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr::iterationStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr::requestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", "tensorrt_llm::executor::JsonSerialization::toJsonStr::requestStatsPerIter"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedDataE", "tensorrt_llm::executor::KVCacheCreatedData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedData22numBlocksPerCacheLevelE", "tensorrt_llm::executor::KVCacheCreatedData::numBlocksPerCacheLevel"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEventE", "tensorrt_llm::executor::KVCacheEvent"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent::eventId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent4dataE", "tensorrt_llm::executor::KVCacheEvent::data"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent7eventIdE", "tensorrt_llm::executor::KVCacheEvent::eventId"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDataE", "tensorrt_llm::executor::KVCacheEventData"], [0, 2, 1, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", "tensorrt_llm::executor::KVCacheEventDiff"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", "tensorrt_llm::executor::KVCacheEventDiff::T"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8newValueE", "tensorrt_llm::executor::KVCacheEventDiff::newValue"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8oldValueE", "tensorrt_llm::executor::KVCacheEventDiff::oldValue"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManagerE", "tensorrt_llm::executor::KVCacheEventManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", "tensorrt_llm::executor::KVCacheEventManager::KVCacheEventManager"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", "tensorrt_llm::executor::KVCacheEventManager::KVCacheEventManager::kvCacheManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KVCacheEventManager::getLatestEvents"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KVCacheEventManager::getLatestEvents::timeout"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager14kvCacheManagerE", "tensorrt_llm::executor::KVCacheEventManager::kvCacheManager"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedDataE", "tensorrt_llm::executor::KVCacheRemovedData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedData11blockHashesE", "tensorrt_llm::executor::KVCacheRemovedData::blockHashes"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockDataE", "tensorrt_llm::executor::KVCacheStoredBlockData"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::blockHash"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::cacheLevel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::loraId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::tokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData9blockHashE", "tensorrt_llm::executor::KVCacheStoredBlockData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData10cacheLevelE", "tensorrt_llm::executor::KVCacheStoredBlockData::cacheLevel"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6loraIdE", "tensorrt_llm::executor::KVCacheStoredBlockData::loraId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData8priorityE", "tensorrt_llm::executor::KVCacheStoredBlockData::priority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6tokensE", "tensorrt_llm::executor::KVCacheStoredBlockData::tokens"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredDataE", "tensorrt_llm::executor::KVCacheStoredData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData6blocksE", "tensorrt_llm::executor::KVCacheStoredData::blocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData10parentHashE", "tensorrt_llm::executor::KVCacheStoredData::parentHash"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedDataE", "tensorrt_llm::executor::KVCacheUpdatedData"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", "tensorrt_llm::executor::KVCacheUpdatedData::KVCacheUpdatedData"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", "tensorrt_llm::executor::KVCacheUpdatedData::KVCacheUpdatedData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData9blockHashE", "tensorrt_llm::executor::KVCacheUpdatedData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData10cacheLevelE", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevel"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated::newValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated::oldValue"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData8priorityE", "tensorrt_llm::executor::KVCacheUpdatedData::priority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated::newValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated::oldValue"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfigE", "tensorrt_llm::executor::KvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::copyOnPartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::crossKvCacheFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::enableBlockReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::enablePartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::eventBufferMaxSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::freeGpuMemoryFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::hostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::maxAttentionWindowVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::maxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::onboardBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::runtimeDefaults"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::secondaryOffloadMinPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::sinkTokenLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::executor::KvCacheConfig::fillEmptyFieldsFromRuntimeDefaults"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::executor::KvCacheConfig::fillEmptyFieldsFromRuntimeDefaults::runtimeDefaults"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getCopyOnPartialReuseEv", "tensorrt_llm::executor::KvCacheConfig::getCopyOnPartialReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig23getCrossKvCacheFractionEv", "tensorrt_llm::executor::KvCacheConfig::getCrossKvCacheFraction"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig19getEnableBlockReuseEv", "tensorrt_llm::executor::KvCacheConfig::getEnableBlockReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEnablePartialReuseEv", "tensorrt_llm::executor::KvCacheConfig::getEnablePartialReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEventBufferMaxSizeEv", "tensorrt_llm::executor::KvCacheConfig::getEventBufferMaxSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getFreeGpuMemoryFractionEv", "tensorrt_llm::executor::KvCacheConfig::getFreeGpuMemoryFraction"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getHostCacheSizeEv", "tensorrt_llm::executor::KvCacheConfig::getHostCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getMaxAttentionWindowVecEv", "tensorrt_llm::executor::KvCacheConfig::getMaxAttentionWindowVec"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig12getMaxTokensEv", "tensorrt_llm::executor::KvCacheConfig::getMaxTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getOnboardBlocksEv", "tensorrt_llm::executor::KvCacheConfig::getOnboardBlocks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig30getSecondaryOffloadMinPriorityEv", "tensorrt_llm::executor::KvCacheConfig::getSecondaryOffloadMinPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig18getSinkTokenLengthEv", "tensorrt_llm::executor::KvCacheConfig::getSinkTokenLength"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mCopyOnPartialReuseE", "tensorrt_llm::executor::KvCacheConfig::mCopyOnPartialReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21mCrossKvCacheFractionE", "tensorrt_llm::executor::KvCacheConfig::mCrossKvCacheFraction"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig17mEnableBlockReuseE", "tensorrt_llm::executor::KvCacheConfig::mEnableBlockReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEnablePartialReuseE", "tensorrt_llm::executor::KvCacheConfig::mEnablePartialReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEventBufferMaxSizeE", "tensorrt_llm::executor::KvCacheConfig::mEventBufferMaxSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mFreeGpuMemoryFractionE", "tensorrt_llm::executor::KvCacheConfig::mFreeGpuMemoryFraction"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mHostCacheSizeE", "tensorrt_llm::executor::KvCacheConfig::mHostCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mMaxAttentionWindowVecE", "tensorrt_llm::executor::KvCacheConfig::mMaxAttentionWindowVec"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig10mMaxTokensE", "tensorrt_llm::executor::KvCacheConfig::mMaxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mOnboardBlocksE", "tensorrt_llm::executor::KvCacheConfig::mOnboardBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig28mSecondaryOffloadMinPriorityE", "tensorrt_llm::executor::KvCacheConfig::mSecondaryOffloadMinPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16mSinkTokenLengthE", "tensorrt_llm::executor::KvCacheConfig::mSinkTokenLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setCopyOnPartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setCopyOnPartialReuse::copyOnPartialReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setCrossKvCacheFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setCrossKvCacheFraction::crossKvCacheFraction"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnableBlockReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnableBlockReuse::enableBlockReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnablePartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnablePartialReuse::enablePartialReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setEventBufferMaxSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setEventBufferMaxSize::eventBufferMaxSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setFreeGpuMemoryFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setFreeGpuMemoryFraction::freeGpuMemoryFraction"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setHostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setHostCacheSize::hostCacheSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", "tensorrt_llm::executor::KvCacheConfig::setMaxAttentionWindowVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", "tensorrt_llm::executor::KvCacheConfig::setMaxAttentionWindowVec::maxAttentionWindowVec"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setMaxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setMaxTokens::maxTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", "tensorrt_llm::executor::KvCacheConfig::setOnboardBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", "tensorrt_llm::executor::KvCacheConfig::setOnboardBlocks::onboardBlocks"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", "tensorrt_llm::executor::KvCacheConfig::setSecondaryOffloadMinPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", "tensorrt_llm::executor::KvCacheConfig::setSecondaryOffloadMinPriority::secondaryOffloadMinPriority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setSinkTokenLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setSinkTokenLength::sinkTokenLength"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfigE", "tensorrt_llm::executor::KvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigEv", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::decodeDurationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::decodeRetentionPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::tokenRangeRetentionPriorities"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::durationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::tokenEnd"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::tokenStart"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10durationMsE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::durationMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::operator==::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8priorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::priority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8tokenEndE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::tokenEnd"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10tokenStartE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::tokenStart"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig19getDecodeDurationMsEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeDurationMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig26getDecodeRetentionPriorityEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeRetentionPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration::blockSize"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration::seqLen"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig29getTokenRangeRetentionConfigsEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getTokenRangeRetentionConfigs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25kDefaultRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kDefaultRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMaxRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kMaxRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMinRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kMinRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig17mDecodeDurationMsE", "tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeDurationMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig24mDecodeRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig27mTokenRangeRetentionConfigsE", "tensorrt_llm::executor::KvCacheRetentionConfig::mTokenRangeRetentionConfigs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStatsE", "tensorrt_llm::executor::KvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14allocNewBlocksE", "tensorrt_llm::executor::KvCacheStats::allocNewBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats16allocTotalBlocksE", "tensorrt_llm::executor::KvCacheStats::allocTotalBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12cacheHitRateE", "tensorrt_llm::executor::KvCacheStats::cacheHitRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13freeNumBlocksE", "tensorrt_llm::executor::KvCacheStats::freeNumBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12maxNumBlocksE", "tensorrt_llm::executor::KvCacheStats::maxNumBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12missedBlocksE", "tensorrt_llm::executor::KvCacheStats::missedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12reusedBlocksE", "tensorrt_llm::executor::KvCacheStats::reusedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14tokensPerBlockE", "tensorrt_llm::executor::KvCacheStats::tokensPerBlock"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13usedNumBlocksE", "tensorrt_llm::executor::KvCacheStats::usedNumBlocks"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor19LogitsPostProcessorE", "tensorrt_llm::executor::LogitsPostProcessor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor26LogitsPostProcessorBatchedE", "tensorrt_llm::executor::LogitsPostProcessorBatched"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfigE", "tensorrt_llm::executor::LogitsPostProcessorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::processorBatched"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::processorMap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::replicate"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig19getProcessorBatchedEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getProcessorBatched"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig15getProcessorMapEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getProcessorMap"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig12getReplicateEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getReplicate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig17mProcessorBatchedE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mProcessorBatched"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig13mProcessorMapE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mProcessorMap"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig10mReplicateE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mReplicate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorBatched"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorBatched::processorBatched"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorMap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorMap::processorMap"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::setReplicate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::setReplicate::replicate"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor22LogitsPostProcessorMapE", "tensorrt_llm::executor::LogitsPostProcessorMap"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfigE", "tensorrt_llm::executor::LookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigEv", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::windowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig28calculateSpeculativeResourceEv", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResource"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::windowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig3getEv", "tensorrt_llm::executor::LookaheadDecodingConfig::get"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig12getNgramSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig22getVerificationSetSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getVerificationSetSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig13getWindowSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getWindowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::isLE"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::isLE::that"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::windowSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig30kDefaultLookaheadDecodingNgramE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingNgram"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig40kDefaultLookaheadDecodingVerificationSetE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingVerificationSet"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig31kDefaultLookaheadDecodingWindowE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig10mNgramSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig20mVerificationSetSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mVerificationSetSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig11mWindowSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mWindowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfigE", "tensorrt_llm::executor::LoraConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::taskId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::weights"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getConfigEv", "tensorrt_llm::executor::LoraConfig::getConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getTaskIdEv", "tensorrt_llm::executor::LoraConfig::getTaskId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig10getWeightsEv", "tensorrt_llm::executor::LoraConfig::getWeights"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mConfigE", "tensorrt_llm::executor::LoraConfig::mConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mTaskIdE", "tensorrt_llm::executor::LoraConfig::mTaskId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig8mWeightsE", "tensorrt_llm::executor::LoraConfig::mWeights"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor13MedusaChoicesE", "tensorrt_llm::executor::MedusaChoices"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor10MemoryTypeE", "tensorrt_llm::executor::MemoryType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kCPUE", "tensorrt_llm::executor::MemoryType::kCPU"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType11kCPU_PINNEDE", "tensorrt_llm::executor::MemoryType::kCPU_PINNED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType15kCPU_PINNEDPOOLE", "tensorrt_llm::executor::MemoryType::kCPU_PINNEDPOOL"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kGPUE", "tensorrt_llm::executor::MemoryType::kGPU"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType8kUNKNOWNE", "tensorrt_llm::executor::MemoryType::kUNKNOWN"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kUVME", "tensorrt_llm::executor::MemoryType::kUVM"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16MillisecondsTypeE", "tensorrt_llm::executor::MillisecondsType"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor9ModelTypeE", "tensorrt_llm::executor::ModelType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType13kDECODER_ONLYE", "tensorrt_llm::executor::ModelType::kDECODER_ONLY"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType16kENCODER_DECODERE", "tensorrt_llm::executor::ModelType::kENCODER_DECODER"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType13kENCODER_ONLYE", "tensorrt_llm::executor::ModelType::kENCODER_ONLY"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfigE", "tensorrt_llm::executor::MropeConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig::mropePositionDeltas"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig::mropeRoratySinCos"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11MropeConfig22getMRopePositionDeltasEv", "tensorrt_llm::executor::MropeConfig::getMRopePositionDeltas"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11MropeConfig20getMRopeRotaryCosSinEv", "tensorrt_llm::executor::MropeConfig::getMRopeRotaryCosSin"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig20mMRopePositionDeltasE", "tensorrt_llm::executor::MropeConfig::mMRopePositionDeltas"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig18mMRopeRotaryCosSinE", "tensorrt_llm::executor::MropeConfig::mMRopeRotaryCosSin"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfigE", "tensorrt_llm::executor::OrchestratorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::isOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::orchLeaderComm"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::spawnProcesses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::workerExecutablePath"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getIsOrchestratorEv", "tensorrt_llm::executor::OrchestratorConfig::getIsOrchestrator"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getOrchLeaderCommEv", "tensorrt_llm::executor::OrchestratorConfig::getOrchLeaderComm"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getSpawnProcessesEv", "tensorrt_llm::executor::OrchestratorConfig::getSpawnProcesses"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig23getWorkerExecutablePathEv", "tensorrt_llm::executor::OrchestratorConfig::getWorkerExecutablePath"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mIsOrchestratorE", "tensorrt_llm::executor::OrchestratorConfig::mIsOrchestrator"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mOrchLeaderCommE", "tensorrt_llm::executor::OrchestratorConfig::mOrchLeaderComm"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mSpawnProcessesE", "tensorrt_llm::executor::OrchestratorConfig::mSpawnProcesses"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig21mWorkerExecutablePathE", "tensorrt_llm::executor::OrchestratorConfig::mWorkerExecutablePath"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", "tensorrt_llm::executor::OrchestratorConfig::setIsOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", "tensorrt_llm::executor::OrchestratorConfig::setIsOrchestrator::isOrchestrator"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", "tensorrt_llm::executor::OrchestratorConfig::setOrchLeaderComm"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", "tensorrt_llm::executor::OrchestratorConfig::setOrchLeaderComm::orchLeaderComm"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", "tensorrt_llm::executor::OrchestratorConfig::setSpawnProcesses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", "tensorrt_llm::executor::OrchestratorConfig::setSpawnProcesses::spawnProcesses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", "tensorrt_llm::executor::OrchestratorConfig::setWorkerExecutablePath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", "tensorrt_llm::executor::OrchestratorConfig::setWorkerExecutablePath::workerExecutablePath"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfigE", "tensorrt_llm::executor::OutputConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::additionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::excludeInputFromOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnContextLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnEncoderOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnPerfMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22additionalModelOutputsE", "tensorrt_llm::executor::OutputConfig::additionalModelOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22excludeInputFromOutputE", "tensorrt_llm::executor::OutputConfig::excludeInputFromOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnContextLogitsE", "tensorrt_llm::executor::OutputConfig::returnContextLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnEncoderOutputE", "tensorrt_llm::executor::OutputConfig::returnEncoderOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22returnGenerationLogitsE", "tensorrt_llm::executor::OutputConfig::returnGenerationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig14returnLogProbsE", "tensorrt_llm::executor::OutputConfig::returnLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig17returnPerfMetricsE", "tensorrt_llm::executor::OutputConfig::returnPerfMetrics"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfigE", "tensorrt_llm::executor::ParallelConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::commMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::commType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::deviceIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::numNodes"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::participantIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationModeEv", "tensorrt_llm::executor::ParallelConfig::getCommunicationMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationTypeEv", "tensorrt_llm::executor::ParallelConfig::getCommunicationType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig12getDeviceIdsEv", "tensorrt_llm::executor::ParallelConfig::getDeviceIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig11getNumNodesEv", "tensorrt_llm::executor::ParallelConfig::getNumNodes"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig21getOrchestratorConfigEv", "tensorrt_llm::executor::ParallelConfig::getOrchestratorConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig17getParticipantIdsEv", "tensorrt_llm::executor::ParallelConfig::getParticipantIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommModeE", "tensorrt_llm::executor::ParallelConfig::mCommMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommTypeE", "tensorrt_llm::executor::ParallelConfig::mCommType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig10mDeviceIdsE", "tensorrt_llm::executor::ParallelConfig::mDeviceIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mNumNodesE", "tensorrt_llm::executor::ParallelConfig::mNumNodes"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig19mOrchestratorConfigE", "tensorrt_llm::executor::ParallelConfig::mOrchestratorConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig15mParticipantIdsE", "tensorrt_llm::executor::ParallelConfig::mParticipantIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", "tensorrt_llm::executor::ParallelConfig::setCommunicationMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", "tensorrt_llm::executor::ParallelConfig::setCommunicationMode::mode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", "tensorrt_llm::executor::ParallelConfig::setCommunicationType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", "tensorrt_llm::executor::ParallelConfig::setCommunicationType::type"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setDeviceIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setDeviceIds::deviceIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", "tensorrt_llm::executor::ParallelConfig::setNumNodes"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", "tensorrt_llm::executor::ParallelConfig::setNumNodes::numNodes"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", "tensorrt_llm::executor::ParallelConfig::setOrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", "tensorrt_llm::executor::ParallelConfig::setOrchestratorConfig::orchestratorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setParticipantIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setParticipantIds::participantIds"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfigE", "tensorrt_llm::executor::PeftCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::deviceCachePercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::hostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::loraPrefetchDir"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxAdapterSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxPagesPerBlockDevice"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxPagesPerBlockHost"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numCopyStreams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numDeviceModuleLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numEnsureWorkers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numHostModuleLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numPutWorkers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::optimalAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getDeviceCachePercentEv", "tensorrt_llm::executor::PeftCacheConfig::getDeviceCachePercent"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getHostCacheSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getHostCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig18getLoraPrefetchDirEv", "tensorrt_llm::executor::PeftCacheConfig::getLoraPrefetchDir"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getMaxAdapterSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig25getMaxPagesPerBlockDeviceEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxPagesPerBlockDevice"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getMaxPagesPerBlockHostEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxPagesPerBlockHost"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getNumCopyStreamsEv", "tensorrt_llm::executor::PeftCacheConfig::getNumCopyStreams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getNumDeviceModuleLayerEv", "tensorrt_llm::executor::PeftCacheConfig::getNumDeviceModuleLayer"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig19getNumEnsureWorkersEv", "tensorrt_llm::executor::PeftCacheConfig::getNumEnsureWorkers"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getNumHostModuleLayerEv", "tensorrt_llm::executor::PeftCacheConfig::getNumHostModuleLayer"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getNumPutWorkersEv", "tensorrt_llm::executor::PeftCacheConfig::getNumPutWorkers"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getOptimalAdapterSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getOptimalAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig22kDefaultMaxAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig30kDefaultMaxPagesPerBlockDeviceE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxPagesPerBlockDevice"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig28kDefaultMaxPagesPerBlockHostE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxPagesPerBlockHost"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig26kDefaultOptimalAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultOptimalAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mDeviceCachePercentE", "tensorrt_llm::executor::PeftCacheConfig::mDeviceCachePercent"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mHostCacheSizeE", "tensorrt_llm::executor::PeftCacheConfig::mHostCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig16mLoraPrefetchDirE", "tensorrt_llm::executor::PeftCacheConfig::mLoraPrefetchDir"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mMaxAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::mMaxAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig23mMaxPagesPerBlockDeviceE", "tensorrt_llm::executor::PeftCacheConfig::mMaxPagesPerBlockDevice"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mMaxPagesPerBlockHostE", "tensorrt_llm::executor::PeftCacheConfig::mMaxPagesPerBlockHost"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mNumCopyStreamsE", "tensorrt_llm::executor::PeftCacheConfig::mNumCopyStreams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mNumDeviceModuleLayerE", "tensorrt_llm::executor::PeftCacheConfig::mNumDeviceModuleLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig17mNumEnsureWorkersE", "tensorrt_llm::executor::PeftCacheConfig::mNumEnsureWorkers"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mNumHostModuleLayerE", "tensorrt_llm::executor::PeftCacheConfig::mNumHostModuleLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mNumPutWorkersE", "tensorrt_llm::executor::PeftCacheConfig::mNumPutWorkers"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mOptimalAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::mOptimalAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", "tensorrt_llm::executor::PeftCacheConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", "tensorrt_llm::executor::PeftCacheConfig::operator==::other"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12PriorityTypeE", "tensorrt_llm::executor::PriorityType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfigE", "tensorrt_llm::executor::PromptTuningConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig::embeddingTable"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig::inputTokenExtraIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig17getEmbeddingTableEv", "tensorrt_llm::executor::PromptTuningConfig::getEmbeddingTable"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig21getInputTokenExtraIdsEv", "tensorrt_llm::executor::PromptTuningConfig::getInputTokenExtraIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig15mEmbeddingTableE", "tensorrt_llm::executor::PromptTuningConfig::mEmbeddingTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig19mInputTokenExtraIdsE", "tensorrt_llm::executor::PromptTuningConfig::mInputTokenExtraIds"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor14RandomSeedTypeE", "tensorrt_llm::executor::RandomSeedType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor7RequestE", "tensorrt_llm::executor::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", "tensorrt_llm::executor::Request::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", "tensorrt_llm::executor::Request::Request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::allottedTimeMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::badWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::crossAttentionMask"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::embeddingBias"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderInputFeatures"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderInputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderOutputLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::endId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::externalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::inputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::languageAdapterUid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::logitsPostProcessor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::logitsPostProcessorName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::lookaheadConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::loraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::mRopeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::maxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::multimodalEmbedding"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::numReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", "tensorrt_llm::executor::Request::Request::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", "tensorrt_llm::executor::Request::Request::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::outputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::pTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::padId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::positionIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::returnAllGeneratedTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::samplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::skipCrossAttnBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::stopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::streaming"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::type"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request24getAdditionalOutputNamesEv", "tensorrt_llm::executor::Request::getAdditionalOutputNames"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request17getAllottedTimeMsEv", "tensorrt_llm::executor::Request::getAllottedTimeMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getBadWordsEv", "tensorrt_llm::executor::Request::getBadWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getClientIdEv", "tensorrt_llm::executor::Request::getClientId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getContextPhaseParamsEv", "tensorrt_llm::executor::Request::getContextPhaseParams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getCrossAttentionMaskEv", "tensorrt_llm::executor::Request::getCrossAttentionMask"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getEagleConfigEv", "tensorrt_llm::executor::Request::getEagleConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request16getEmbeddingBiasEv", "tensorrt_llm::executor::Request::getEmbeddingBias"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputFeaturesEv", "tensorrt_llm::executor::Request::getEncoderInputFeatures"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputTokenIdsEv", "tensorrt_llm::executor::Request::getEncoderInputTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getEncoderOutputLengthEv", "tensorrt_llm::executor::Request::getEncoderOutputLength"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request8getEndIdEv", "tensorrt_llm::executor::Request::getEndId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request28getExternalDraftTokensConfigEv", "tensorrt_llm::executor::Request::getExternalDraftTokensConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getGuidedDecodingParamsEv", "tensorrt_llm::executor::Request::getGuidedDecodingParams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request16getInputTokenIdsEv", "tensorrt_llm::executor::Request::getInputTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request25getKvCacheRetentionConfigEv", "tensorrt_llm::executor::Request::getKvCacheRetentionConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getLanguageAdapterUidEv", "tensorrt_llm::executor::Request::getLanguageAdapterUid"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getLogitsPostProcessorEv", "tensorrt_llm::executor::Request::getLogitsPostProcessor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request26getLogitsPostProcessorNameEv", "tensorrt_llm::executor::Request::getLogitsPostProcessorName"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request18getLookaheadConfigEv", "tensorrt_llm::executor::Request::getLookaheadConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request13getLoraConfigEv", "tensorrt_llm::executor::Request::getLoraConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getMaxTokensEv", "tensorrt_llm::executor::Request::getMaxTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getMropeConfigEv", "tensorrt_llm::executor::Request::getMropeConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getMultimodalEmbeddingEv", "tensorrt_llm::executor::Request::getMultimodalEmbedding"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request15getOutputConfigEv", "tensorrt_llm::executor::Request::getOutputConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request8getPadIdEv", "tensorrt_llm::executor::Request::getPadId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getPositionIdsEv", "tensorrt_llm::executor::Request::getPositionIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getPriorityEv", "tensorrt_llm::executor::Request::getPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getPromptTuningConfigEv", "tensorrt_llm::executor::Request::getPromptTuningConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getRequestTypeEv", "tensorrt_llm::executor::Request::getRequestType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request27getReturnAllGeneratedTokensEv", "tensorrt_llm::executor::Request::getReturnAllGeneratedTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request17getSamplingConfigEv", "tensorrt_llm::executor::Request::getSamplingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getSkipCrossAttnBlocksEv", "tensorrt_llm::executor::Request::getSkipCrossAttnBlocks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getStopWordsEv", "tensorrt_llm::executor::Request::getStopWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getStreamingEv", "tensorrt_llm::executor::Request::getStreaming"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request25kBatchedPostProcessorNameE", "tensorrt_llm::executor::Request::kBatchedPostProcessorName"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request16kDefaultPriorityE", "tensorrt_llm::executor::Request::kDefaultPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request31kDynamicPostProcessorNamePrefixE", "tensorrt_llm::executor::Request::kDynamicPostProcessorNamePrefix"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request5mImplE", "tensorrt_llm::executor::Request::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", "tensorrt_llm::executor::Request::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", "tensorrt_llm::executor::Request::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", "tensorrt_llm::executor::Request::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", "tensorrt_llm::executor::Request::operator=::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", "tensorrt_llm::executor::Request::setAllottedTimeMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", "tensorrt_llm::executor::Request::setAllottedTimeMs::allottedTimeMs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setBadWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setBadWords::badWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", "tensorrt_llm::executor::Request::setClientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", "tensorrt_llm::executor::Request::setClientId::clientId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", "tensorrt_llm::executor::Request::setContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", "tensorrt_llm::executor::Request::setContextPhaseParams::contextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", "tensorrt_llm::executor::Request::setCrossAttentionMask"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", "tensorrt_llm::executor::Request::setCrossAttentionMask::crossAttentionMask"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::Request::setEagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::Request::setEagleConfig::eagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", "tensorrt_llm::executor::Request::setEmbeddingBias"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", "tensorrt_llm::executor::Request::setEmbeddingBias::embeddingBias"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", "tensorrt_llm::executor::Request::setEncoderInputFeatures"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", "tensorrt_llm::executor::Request::setEncoderInputFeatures::encoderInputFeatures"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", "tensorrt_llm::executor::Request::setEncoderInputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", "tensorrt_llm::executor::Request::setEncoderInputTokenIds::encoderInputTokenIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", "tensorrt_llm::executor::Request::setEncoderOutputLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", "tensorrt_llm::executor::Request::setEncoderOutputLength::encoderOutputLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", "tensorrt_llm::executor::Request::setEndId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", "tensorrt_llm::executor::Request::setEndId::endId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Request::setExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Request::setExternalDraftTokensConfig::externalDraftTokensConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", "tensorrt_llm::executor::Request::setGuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", "tensorrt_llm::executor::Request::setGuidedDecodingParams::guidedDecodingParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Request::setKvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Request::setKvCacheRetentionConfig::kvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", "tensorrt_llm::executor::Request::setLanguageAdapterUid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", "tensorrt_llm::executor::Request::setLanguageAdapterUid::languageAdapterUid"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", "tensorrt_llm::executor::Request::setLogitsPostProcessor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", "tensorrt_llm::executor::Request::setLogitsPostProcessor::logitsPostProcessor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", "tensorrt_llm::executor::Request::setLogitsPostProcessorName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", "tensorrt_llm::executor::Request::setLogitsPostProcessorName::logitsPostProcessorName"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Request::setLookaheadConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Request::setLookaheadConfig::lookaheadConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", "tensorrt_llm::executor::Request::setLoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", "tensorrt_llm::executor::Request::setLoraConfig::loraConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", "tensorrt_llm::executor::Request::setMropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", "tensorrt_llm::executor::Request::setMropeConfig::mRopeConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", "tensorrt_llm::executor::Request::setMultimodalEmbedding"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", "tensorrt_llm::executor::Request::setMultimodalEmbedding::multimodalEmbedding"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", "tensorrt_llm::executor::Request::setOutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", "tensorrt_llm::executor::Request::setOutputConfig::outputConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", "tensorrt_llm::executor::Request::setPadId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", "tensorrt_llm::executor::Request::setPadId::padId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::Request::setPositionIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::Request::setPositionIds::positionIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", "tensorrt_llm::executor::Request::setPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", "tensorrt_llm::executor::Request::setPriority::priority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", "tensorrt_llm::executor::Request::setPromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", "tensorrt_llm::executor::Request::setPromptTuningConfig::pTuningConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", "tensorrt_llm::executor::Request::setRequestType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", "tensorrt_llm::executor::Request::setRequestType::requestType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", "tensorrt_llm::executor::Request::setReturnAllGeneratedTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", "tensorrt_llm::executor::Request::setReturnAllGeneratedTokens::returnAllGeneratedTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", "tensorrt_llm::executor::Request::setSamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", "tensorrt_llm::executor::Request::setSamplingConfig::config"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", "tensorrt_llm::executor::Request::setSkipCrossAttnBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", "tensorrt_llm::executor::Request::setSkipCrossAttnBlocks::skipCrossAttnBlocks"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setStopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setStopWords::stopWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", "tensorrt_llm::executor::Request::setStreaming"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", "tensorrt_llm::executor::Request::setStreaming::streaming"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestD0Ev", "tensorrt_llm::executor::Request::~Request"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetricsE", "tensorrt_llm::executor::RequestPerfMetrics"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics14kvCacheHitRateE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::kvCacheHitRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numMissedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numMissedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics21numNewAllocatedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numNewAllocatedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numReusedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numReusedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics23numTotalAllocatedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numTotalAllocatedBlocks"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics14acceptanceRateE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::acceptanceRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics24totalAcceptedDraftTokensE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::totalAcceptedDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics16totalDraftTokensE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::totalDraftTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9TimePointE", "tensorrt_llm::executor::RequestPerfMetrics::TimePoint"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11arrivalTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::arrivalTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18firstScheduledTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::firstScheduledTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics14firstTokenTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::firstTokenTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11kvCacheSizeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18kvCacheTransferEndE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheTransferEnd"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics20kvCacheTransferStartE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheTransferStart"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics13lastTokenTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::lastTokenTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9firstIterE", "tensorrt_llm::executor::RequestPerfMetrics::firstIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics4iterE", "tensorrt_llm::executor::RequestPerfMetrics::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14kvCacheMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::kvCacheMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics8lastIterE", "tensorrt_llm::executor::RequestPerfMetrics::lastIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics19speculativeDecodingE", "tensorrt_llm::executor::RequestPerfMetrics::speculativeDecoding"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13timingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::timingMetrics"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12RequestStageE", "tensorrt_llm::executor::RequestStage"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kCONTEXT_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kCONTEXT_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kENCODER_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kENCODER_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kGENERATION_COMPLETEE", "tensorrt_llm::executor::RequestStage::kGENERATION_COMPLETE"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage23kGENERATION_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kGENERATION_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage7kQUEUEDE", "tensorrt_llm::executor::RequestStage::kQUEUED"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12RequestStatsE", "tensorrt_llm::executor::RequestStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats24allocNewBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::allocNewBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats26allocTotalBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::allocTotalBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats26avgNumDecodedTokensPerIterE", "tensorrt_llm::executor::RequestStats::avgNumDecodedTokensPerIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22contextPrefillPositionE", "tensorrt_llm::executor::RequestStats::contextPrefillPosition"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats15disServingStatsE", "tensorrt_llm::executor::RequestStats::disServingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats2idE", "tensorrt_llm::executor::RequestStats::id"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats24kvCacheHitRatePerRequestE", "tensorrt_llm::executor::RequestStats::kvCacheHitRatePerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22missedBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::missedBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats18numGeneratedTokensE", "tensorrt_llm::executor::RequestStats::numGeneratedTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats6pausedE", "tensorrt_llm::executor::RequestStats::paused"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22reusedBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::reusedBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats9scheduledE", "tensorrt_llm::executor::RequestStats::scheduled"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats5stageE", "tensorrt_llm::executor::RequestStats::stage"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIterationE", "tensorrt_llm::executor::RequestStatsPerIteration"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration4iterE", "tensorrt_llm::executor::RequestStatsPerIteration::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration12requestStatsE", "tensorrt_llm::executor::RequestStatsPerIteration::requestStats"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor11RequestTypeE", "tensorrt_llm::executor::RequestType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType35REQUEST_TYPE_CONTEXT_AND_GENERATIONE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_CONTEXT_AND_GENERATION"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType25REQUEST_TYPE_CONTEXT_ONLYE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_CONTEXT_ONLY"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType28REQUEST_TYPE_GENERATION_ONLYE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_GENERATION_ONLY"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8ResponseE", "tensorrt_llm::executor::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", "tensorrt_llm::executor::Response::Response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::Result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::errorMsg"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", "tensorrt_llm::executor::Response::Response::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", "tensorrt_llm::executor::Response::Response::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::requestId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::requestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response11getClientIdEv", "tensorrt_llm::executor::Response::getClientId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response11getErrorMsgEv", "tensorrt_llm::executor::Response::getErrorMsg"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response12getRequestIdEv", "tensorrt_llm::executor::Response::getRequestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response9getResultEv", "tensorrt_llm::executor::Response::getResult"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response8hasErrorEv", "tensorrt_llm::executor::Response::hasError"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8Response5mImplE", "tensorrt_llm::executor::Response::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", "tensorrt_llm::executor::Response::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", "tensorrt_llm::executor::Response::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", "tensorrt_llm::executor::Response::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", "tensorrt_llm::executor::Response::operator=::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseD0Ev", "tensorrt_llm::executor::Response::~Response"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor6ResultE", "tensorrt_llm::executor::Result"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result17additionalOutputsE", "tensorrt_llm::executor::Result::additionalOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13contextLogitsE", "tensorrt_llm::executor::Result::contextLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result18contextPhaseParamsE", "tensorrt_llm::executor::Result::contextPhaseParams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result11cumLogProbsE", "tensorrt_llm::executor::Result::cumLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result12decodingIterE", "tensorrt_llm::executor::Result::decodingIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13encoderOutputE", "tensorrt_llm::executor::Result::encoderOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13finishReasonsE", "tensorrt_llm::executor::Result::finishReasons"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result16generationLogitsE", "tensorrt_llm::executor::Result::generationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result7isFinalE", "tensorrt_llm::executor::Result::isFinal"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result15isSequenceFinalE", "tensorrt_llm::executor::Result::isSequenceFinal"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result8logProbsE", "tensorrt_llm::executor::Result::logProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result14outputTokenIdsE", "tensorrt_llm::executor::Result::outputTokenIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result18requestPerfMetricsE", "tensorrt_llm::executor::Result::requestPerfMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13sequenceIndexE", "tensorrt_llm::executor::Result::sequenceIndex"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result21specDecFastLogitsInfoE", "tensorrt_llm::executor::Result::specDecFastLogitsInfo"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor17RetentionPriorityE", "tensorrt_llm::executor::RetentionPriority"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDurationE", "tensorrt_llm::executor::RetentionPriorityAndDuration"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration::durationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration::retentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration10durationMsE", "tensorrt_llm::executor::RetentionPriorityAndDuration::durationMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration17retentionPriorityE", "tensorrt_llm::executor::RetentionPriorityAndDuration::retentionPriority"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfigE", "tensorrt_llm::executor::SamplingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::earlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::frequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::lengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::minP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::minTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::noRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::numReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::presencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::repetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::seed"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::temperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPResetIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkBeamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkBeamSearchDiversityRate::beamSearchDiversityRate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidth::beamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray::beamWidthArray"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkEarlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkEarlyStopping::earlyStopping"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkLengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkLengthPenalty::lengthPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkMinP::minP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkMinTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkMinTokens::minTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences::numReturnSequences"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkRepetitionPenalty::repetitionpenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTemperature::temperature"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopK::topK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopP::topP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPDecay::topPDecay"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPMin::topPMin"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPResetIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPResetIds::topPResetIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig26getBeamSearchDiversityRateEv", "tensorrt_llm::executor::SamplingConfig::getBeamSearchDiversityRate"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getBeamWidthEv", "tensorrt_llm::executor::SamplingConfig::getBeamWidth"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getBeamWidthArrayEv", "tensorrt_llm::executor::SamplingConfig::getBeamWidthArray"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getEarlyStoppingEv", "tensorrt_llm::executor::SamplingConfig::getEarlyStopping"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig19getFrequencyPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getFrequencyPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getLengthPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getLengthPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getMinPEv", "tensorrt_llm::executor::SamplingConfig::getMinP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getMinTokensEv", "tensorrt_llm::executor::SamplingConfig::getMinTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getNoRepeatNgramSizeEv", "tensorrt_llm::executor::SamplingConfig::getNoRepeatNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getNumReturnBeamsEv", "tensorrt_llm::executor::SamplingConfig::getNumReturnBeams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig21getNumReturnSequencesEv", "tensorrt_llm::executor::SamplingConfig::getNumReturnSequences"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig18getPresencePenaltyEv", "tensorrt_llm::executor::SamplingConfig::getPresencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getRepetitionPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getRepetitionPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getSeedEv", "tensorrt_llm::executor::SamplingConfig::getSeed"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig14getTemperatureEv", "tensorrt_llm::executor::SamplingConfig::getTemperature"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopKEv", "tensorrt_llm::executor::SamplingConfig::getTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopPEv", "tensorrt_llm::executor::SamplingConfig::getTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getTopPDecayEv", "tensorrt_llm::executor::SamplingConfig::getTopPDecay"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig10getTopPMinEv", "tensorrt_llm::executor::SamplingConfig::getTopPMin"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig15getTopPResetIdsEv", "tensorrt_llm::executor::SamplingConfig::getTopPResetIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig24mBeamSearchDiversityRateE", "tensorrt_llm::executor::SamplingConfig::mBeamSearchDiversityRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mBeamWidthE", "tensorrt_llm::executor::SamplingConfig::mBeamWidth"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mBeamWidthArrayE", "tensorrt_llm::executor::SamplingConfig::mBeamWidthArray"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mEarlyStoppingE", "tensorrt_llm::executor::SamplingConfig::mEarlyStopping"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17mFrequencyPenaltyE", "tensorrt_llm::executor::SamplingConfig::mFrequencyPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mLengthPenaltyE", "tensorrt_llm::executor::SamplingConfig::mLengthPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mMinPE", "tensorrt_llm::executor::SamplingConfig::mMinP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mMinTokensE", "tensorrt_llm::executor::SamplingConfig::mMinTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mNoRepeatNgramSizeE", "tensorrt_llm::executor::SamplingConfig::mNoRepeatNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mNumReturnBeamsE", "tensorrt_llm::executor::SamplingConfig::mNumReturnBeams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19mNumReturnSequencesE", "tensorrt_llm::executor::SamplingConfig::mNumReturnSequences"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16mPresencePenaltyE", "tensorrt_llm::executor::SamplingConfig::mPresencePenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mRepetitionPenaltyE", "tensorrt_llm::executor::SamplingConfig::mRepetitionPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mSeedE", "tensorrt_llm::executor::SamplingConfig::mSeed"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12mTemperatureE", "tensorrt_llm::executor::SamplingConfig::mTemperature"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopKE", "tensorrt_llm::executor::SamplingConfig::mTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopPE", "tensorrt_llm::executor::SamplingConfig::mTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mTopPDecayE", "tensorrt_llm::executor::SamplingConfig::mTopPDecay"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig8mTopPMinE", "tensorrt_llm::executor::SamplingConfig::mTopPMin"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig13mTopPResetIdsE", "tensorrt_llm::executor::SamplingConfig::mTopPResetIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::executor::SamplingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::executor::SamplingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setBeamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setBeamSearchDiversityRate::beamSearchDiversityRate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::setBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::setBeamWidth::beamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::setBeamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::setBeamWidthArray::beamWidthArray"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setEarlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setEarlyStopping::earlyStopping"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setFrequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setFrequencyPenalty::frequencyPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setLengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setLengthPenalty::lengthPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setMinP::minP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setMinTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setMinTokens::minTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNumReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNumReturnSequences::numReturnSequences"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setPresencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setPresencePenalty::presencePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setRepetitionPenalty::repetitionPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", "tensorrt_llm::executor::SamplingConfig::setSeed"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", "tensorrt_llm::executor::SamplingConfig::setSeed::seed"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTemperature::temperature"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setTopK::topK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopP::topP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPDecay::topPDecay"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPMin::topPMin"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPResetIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPResetIds::topPResetIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20updateNumReturnBeamsEv", "tensorrt_llm::executor::SamplingConfig::updateNumReturnBeams"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfigE", "tensorrt_llm::executor::SchedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::capacitySchedulerPolicy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::contextChunkingPolicy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::dynamicBatchConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig26getCapacitySchedulerPolicyEv", "tensorrt_llm::executor::SchedulerConfig::getCapacitySchedulerPolicy"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig24getContextChunkingPolicyEv", "tensorrt_llm::executor::SchedulerConfig::getContextChunkingPolicy"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig21getDynamicBatchConfigEv", "tensorrt_llm::executor::SchedulerConfig::getDynamicBatchConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig24mCapacitySchedulerPolicyE", "tensorrt_llm::executor::SchedulerConfig::mCapacitySchedulerPolicy"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig22mContextChunkingPolicyE", "tensorrt_llm::executor::SchedulerConfig::mContextChunkingPolicy"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig19mDynamicBatchConfigE", "tensorrt_llm::executor::SchedulerConfig::mDynamicBatchConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", "tensorrt_llm::executor::SchedulerConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", "tensorrt_llm::executor::SchedulerConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor13SerializationE", "tensorrt_llm::executor::Serialization"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalModelOutput::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalOutput::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeBool"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeBool::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheTransceiverConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCommState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeContextPhaseParams::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDebugConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingMode::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDisServingRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDisServingRequestStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDynamicBatchConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeEagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeEagleConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExecutorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExecutorConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExtendedRuntimePerfKnobConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExternalDraftTokensConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingParams::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeInflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeInflightBatchingStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStats"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeIterationStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStats::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeIterationStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStatsVec::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheRetentionConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLookaheadDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLoraConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeModelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeModelType::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeMropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeMropeConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOrchestratorConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOutputConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeParallelConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePeftCacheConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePromptTuningConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequest::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestPerfMetrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestPerfMetrics::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStage::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIterationVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIterationVec::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResponse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResponse::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeResponses::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResult"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResult::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSamplingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSchedulerConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSocketState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSocketState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecFastLogitsInfo"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecFastLogitsInfo::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpeculativeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpeculativeDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeStaticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeStaticBatchingStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeString"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeString::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTensor::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTimePoint"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTimePoint::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTokenRangeRetentionConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::additionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::additionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serialize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::dynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::inflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::info"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", "tensorrt_llm::executor::Serialization::serialize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", "tensorrt_llm::executor::Serialization::serialize::iterStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::metrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::requestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", "tensorrt_llm::executor::Serialization::serialize::requestStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", "tensorrt_llm::executor::Serialization::serialize::responses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::staticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::stats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tp"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN18RequestPerfMetrics9TimePointE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", "tensorrt_llm::executor::Serialization::serializedSize::additionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", "tensorrt_llm::executor::Serialization::serializedSize::additionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", "tensorrt_llm::executor::Serialization::serializedSize::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", "tensorrt_llm::executor::Serialization::serializedSize::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serializedSize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", "tensorrt_llm::executor::Serialization::serializedSize::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", "tensorrt_llm::executor::Serialization::serializedSize::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", "tensorrt_llm::executor::Serialization::serializedSize::disServingRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", "tensorrt_llm::executor::Serialization::serializedSize::dynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", "tensorrt_llm::executor::Serialization::serializedSize::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", "tensorrt_llm::executor::Serialization::serializedSize::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::Serialization::serializedSize::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", "tensorrt_llm::executor::Serialization::serializedSize::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize::inflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", "tensorrt_llm::executor::Serialization::serializedSize::info"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", "tensorrt_llm::executor::Serialization::serializedSize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", "tensorrt_llm::executor::Serialization::serializedSize::metrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", "tensorrt_llm::executor::Serialization::serializedSize::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", "tensorrt_llm::executor::Serialization::serializedSize::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", "tensorrt_llm::executor::Serialization::serializedSize::request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", "tensorrt_llm::executor::Serialization::serializedSize::requestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", "tensorrt_llm::executor::Serialization::serializedSize::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", "tensorrt_llm::executor::Serialization::serializedSize::result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", "tensorrt_llm::executor::Serialization::serializedSize::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize::staticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", "tensorrt_llm::executor::Serialization::serializedSize::tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::Serialization::serializedSize::tokenRangeRetentionConfig"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor5ShapeE", "tensorrt_llm::executor::Shape"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor5Shape4BaseE", "tensorrt_llm::executor::Shape::Base"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor5Shape9DimType64E", "tensorrt_llm::executor::Shape::DimType64"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", "tensorrt_llm::executor::Shape::Shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEv", "tensorrt_llm::executor::Shape::Shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", "tensorrt_llm::executor::Shape::Shape::dims"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape::size"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10SizeType32E", "tensorrt_llm::executor::SizeType32"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfigE", "tensorrt_llm::executor::SpeculativeDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", "tensorrt_llm::executor::SpeculativeDecodingConfig::SpeculativeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", "tensorrt_llm::executor::SpeculativeDecodingConfig::SpeculativeDecodingConfig::fastLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig10fastLogitsE", "tensorrt_llm::executor::SpeculativeDecodingConfig::fastLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::SpeculativeDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::SpeculativeDecodingConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfoE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo18draftParticipantIdE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::draftParticipantId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo14draftRequestIdE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::draftRequestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo8toTensorEv", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::toTensor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStatsE", "tensorrt_llm::executor::StaticBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats13emptyGenSlotsE", "tensorrt_llm::executor::StaticBatchingStats::emptyGenSlots"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats18numContextRequestsE", "tensorrt_llm::executor::StaticBatchingStats::numContextRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numCtxTokensE", "tensorrt_llm::executor::StaticBatchingStats::numCtxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numGenTokensE", "tensorrt_llm::executor::StaticBatchingStats::numGenTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats20numScheduledRequestsE", "tensorrt_llm::executor::StaticBatchingStats::numScheduledRequests"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9StreamPtrE", "tensorrt_llm::executor::StreamPtr"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor6TensorE", "tensorrt_llm::executor::Tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::CudaStreamPtr"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6Tensor4ImplE", "tensorrt_llm::executor::Tensor::Impl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorEv", "tensorrt_llm::executor::Tensor::Tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", "tensorrt_llm::executor::Tensor::Tensor::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", "tensorrt_llm::executor::Tensor::Tensor::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::Tensor::tensor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo::stream"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo::tensor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToCpu"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToCpu::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToGpu"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToGpu::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToManaged"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToManaged::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPinned"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPinned::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPooledPinned"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPooledPinned::stream"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu::shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::detail::ofITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::detail::ofITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::Tensor::detail::toITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::Tensor::detail::toITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7getDataEv", "tensorrt_llm::executor::Tensor::getData"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor7getDataEv", "tensorrt_llm::executor::Tensor::getData"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor11getDataTypeEv", "tensorrt_llm::executor::Tensor::getDataType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13getMemoryTypeEv", "tensorrt_llm::executor::Tensor::getMemoryType"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", "tensorrt_llm::executor::Tensor::getRuntimeType"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", "tensorrt_llm::executor::Tensor::getRuntimeType::T"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor8getShapeEv", "tensorrt_llm::executor::Tensor::getShape"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor7getSizeEv", "tensorrt_llm::executor::Tensor::getSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor14getSizeInBytesEv", "tensorrt_llm::executor::Tensor::getSizeInBytes"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::shape"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::stream"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::stream"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7mTensorE", "tensorrt_llm::executor::Tensor::mTensor"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed::shape"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::T"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of::T"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::shape"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensorcvbEv", "tensorrt_llm::executor::Tensor::operator bool"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", "tensorrt_llm::executor::Tensor::operator!="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", "tensorrt_llm::executor::Tensor::operator!=::rhs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", "tensorrt_llm::executor::Tensor::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", "tensorrt_llm::executor::Tensor::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", "tensorrt_llm::executor::Tensor::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", "tensorrt_llm::executor::Tensor::operator=::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", "tensorrt_llm::executor::Tensor::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", "tensorrt_llm::executor::Tensor::operator==::rhs"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned::shape"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom::stream"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setZero"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setZero::stream"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensorD0Ev", "tensorrt_llm::executor::Tensor::~Tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9TensorPtrE", "tensorrt_llm::executor::TensorPtr"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11TokenIdTypeE", "tensorrt_llm::executor::TokenIdType"], [0, 2, 1, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", "tensorrt_llm::executor::TypeTraits"], [0, 8, 1, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", "tensorrt_llm::executor::TypeTraits::T"], [0, 2, 1, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", "tensorrt_llm::executor::TypeTraits<T*>"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", "tensorrt_llm::executor::TypeTraits<T*>::T"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIP1TE5valueE", "tensorrt_llm::executor::TypeTraits<T*>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIbEE", "tensorrt_llm::executor::TypeTraits<bool>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIbE5valueE", "tensorrt_llm::executor::TypeTraits<bool>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIfEE", "tensorrt_llm::executor::TypeTraits<float>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIfE5valueE", "tensorrt_llm::executor::TypeTraits<float>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsI4halfEE", "tensorrt_llm::executor::TypeTraits<half>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsI4halfE5valueE", "tensorrt_llm::executor::TypeTraits<half>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int32_tEEE", "tensorrt_llm::executor::TypeTraits<std::int32_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int32_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int32_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int64_tEEE", "tensorrt_llm::executor::TypeTraits<std::int64_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int64_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int64_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt6int8_tEEE", "tensorrt_llm::executor::TypeTraits<std::int8_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt6int8_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int8_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEEE", "tensorrt_llm::executor::TypeTraits<std::uint8_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::uint8_t>::value"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11VecLogProbsE", "tensorrt_llm::executor::VecLogProbs"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16VecTokenExtraIdsE", "tensorrt_llm::executor::VecTokenExtraIds"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9VecTokensE", "tensorrt_llm::executor::VecTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6detailE", "tensorrt_llm::executor::detail"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6detail9DimType64E", "tensorrt_llm::executor::detail::DimType64"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::detail::ofITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::detail::ofITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::detail::toITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::detail::toITensor::tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executorE", "tensorrt_llm::executor::disagg_executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::ctxEnginePaths"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::ctxExecutorConfigs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::genEnginePaths"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::genExecutorConfigs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::hasContextAwaitThreads"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::hasGenAwaitThreads"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses::contextIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses::timeout"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses::genIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses::timeout"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator10canEnqueueEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::canEnqueue"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::batch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::requests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::selectContextId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::batch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::globalRequestIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::requests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::selectGenIdx"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator19getContextExecutorsEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::getContextExecutors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator15getGenExecutorsEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::getGenExecutors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator5mImplE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorD0Ev", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::~DisaggExecutorOrchestrator"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdE", "tensorrt_llm::executor::disagg_executor::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::gid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::gid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::response"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId3gidE", "tensorrt_llm::executor::disagg_executor::ResponseWithId::gid"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator=::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId8responseE", "tensorrt_llm::executor::disagg_executor::ResponseWithId::response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdD0Ev", "tensorrt_llm::executor::disagg_executor::ResponseWithId::~ResponseWithId"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfigE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig::kvFactor"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig14mAttentionTypeE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::mAttentionType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig9mKvFactorE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::mKvFactor"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionTypeE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType8kDEFAULTE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType::kDEFAULT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType4kMLAE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType::kMLA"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPrank"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPrank"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPsize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPsize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::enableAttentionDP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::enableAttentionDP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::modelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbAttentionLayers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbKvHeadPerLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbKvHeads"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::pipelineParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::pipelineParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::sizePerHead"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::sizePerHead"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tensorParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tensorParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tokensPerBlock"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tokensPerBlock"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::worldConfig"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig18mNbKvHeadsPerLayerE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mNbKvHeadsPerLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig12mSizePerHeadE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mSizePerHead"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig15mTokensPerBlockE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mTokensPerBlock"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPrankE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mDPrank"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPsizeE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mDPsize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mEnableAttentionDPE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mEnableAttentionDP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig20mPipelineParallelismE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mPipelineParallelism"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mTensorParallelismE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mTensorParallelism"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState18getAttentionConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getAttentionConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11getDataTypeEv", "tensorrt_llm::executor::kv_cache::CacheState::getDataType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14getModelConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getModelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState17getParallelConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState16mAttentionConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mAttentionConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState9mDataTypeE", "tensorrt_llm::executor::kv_cache::CacheState::mDataType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState12mModelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mModelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15mParallelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mParallelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState8toStringEv", "tensorrt_llm::executor::kv_cache::CacheState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommStateE", "tensorrt_llm::executor::kv_cache::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateEv", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState::ip"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState::port"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::ranks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::selfIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::selfIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::socketState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState11getMpiStateEv", "tensorrt_llm::executor::kv_cache::CommState::getMpiState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10getSelfIdxEv", "tensorrt_llm::executor::kv_cache::CommState::getSelfIdx"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState14getSocketStateEv", "tensorrt_llm::executor::kv_cache::CommState::getSocketState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10isMpiStateEv", "tensorrt_llm::executor::kv_cache::CommState::isMpiState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13isSocketStateEv", "tensorrt_llm::executor::kv_cache::CommState::isSocketState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState8mSelfIdxE", "tensorrt_llm::executor::kv_cache::CommState::mSelfIdx"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState6mStateE", "tensorrt_llm::executor::kv_cache::CommState::mState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", "tensorrt_llm::executor::kv_cache::CommState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", "tensorrt_llm::executor::kv_cache::CommState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState8toStringEv", "tensorrt_llm::executor::kv_cache::CommState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionE", "tensorrt_llm::executor::kv_cache::Connection"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection12isThreadSafeEv", "tensorrt_llm::executor::kv_cache::Connection::isThreadSafe"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::ctx"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::data"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::size"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::ctx"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::data"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::size"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionD0Ev", "tensorrt_llm::executor::kv_cache::Connection::~Connection"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerE", "tensorrt_llm::executor::kv_cache::ConnectionManager"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache17ConnectionManager12getCommStateEv", "tensorrt_llm::executor::kv_cache::ConnectionManager::getCommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", "tensorrt_llm::executor::kv_cache::ConnectionManager::getConnections"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", "tensorrt_llm::executor::kv_cache::ConnectionManager::getConnections::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::ctx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::size"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerD0Ev", "tensorrt_llm::executor::kv_cache::ConnectionManager::~ConnectionManager"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContextE", "tensorrt_llm::executor::kv_cache::DataContext"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", "tensorrt_llm::executor::kv_cache::DataContext::DataContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", "tensorrt_llm::executor::kv_cache::DataContext::DataContext::tag"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11DataContext6getTagEv", "tensorrt_llm::executor::kv_cache::DataContext::getTag"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext4mTagE", "tensorrt_llm::executor::kv_cache::DataContext::mTag"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiStateE", "tensorrt_llm::executor::kv_cache::MpiState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiState6mRanksE", "tensorrt_llm::executor::kv_cache::MpiState::mRanks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", "tensorrt_llm::executor::kv_cache::MpiState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", "tensorrt_llm::executor::kv_cache::MpiState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiState8toStringEv", "tensorrt_llm::executor::kv_cache::MpiState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketStateE", "tensorrt_llm::executor::kv_cache::SocketState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState3mIpE", "tensorrt_llm::executor::kv_cache::SocketState::mIp"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState5mPortE", "tensorrt_llm::executor::kv_cache::SocketState::mPort"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", "tensorrt_llm::executor::kv_cache::SocketState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", "tensorrt_llm::executor::kv_cache::SocketState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketState8toStringEv", "tensorrt_llm::executor::kv_cache::SocketState::toString"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<::policy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<::policy"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7versionEv", "tensorrt_llm::executor::version"], [1, 1, 1, "_CPPv4N12tensorrt_llm6layersE", "tensorrt_llm::layers"], [0, 1, 1, "_CPPv4N12tensorrt_llm3mpiE", "tensorrt_llm::mpi"], [0, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [0, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffersE", "tensorrt_llm::runtime::AllReduceBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::fakeBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9TensorPtrE", "tensorrt_llm::runtime::AllReduceBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers18mAllReduceCommPtrsE", "tensorrt_llm::runtime::AllReduceBuffers::mAllReduceCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers17mIpcMemoryHandlesE", "tensorrt_llm::runtime::AllReduceBuffers::mIpcMemoryHandles"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataTypeE", "tensorrt_llm::runtime::BufferDataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::_unsigned"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::dataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::pointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType11getDataTypeEv", "tensorrt_llm::runtime::BufferDataType::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType7getSizeEv", "tensorrt_llm::runtime::BufferDataType::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType13getSizeInBitsEv", "tensorrt_llm::runtime::BufferDataType::getSizeInBits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType9isPointerEv", "tensorrt_llm::runtime::BufferDataType::isPointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType10isUnsignedEv", "tensorrt_llm::runtime::BufferDataType::isUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType15kTrtPointerTypeE", "tensorrt_llm::runtime::BufferDataType::kTrtPointerType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mDataTypeE", "tensorrt_llm::runtime::BufferDataType::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType8mPointerE", "tensorrt_llm::runtime::BufferDataType::mPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mUnsignedE", "tensorrt_llm::runtime::BufferDataType::mUnsigned"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataTypecvN8nvinfer18DataTypeEEv", "tensorrt_llm::runtime::BufferDataType::operator nvinfer1::DataType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerE", "tensorrt_llm::runtime::BufferManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager::trimPool"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager14CudaMemPoolPtrE", "tensorrt_llm::runtime::BufferManager::CudaMemPoolPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13CudaStreamPtrE", "tensorrt_llm::runtime::BufferManager::CudaStreamPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10IBufferPtrE", "tensorrt_llm::runtime::BufferManager::IBufferPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10ITensorPtrE", "tensorrt_llm::runtime::BufferManager::ITensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::dims"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::size"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::type"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dstType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::srcType"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager9getStreamEv", "tensorrt_llm::runtime::BufferManager::getStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::dims"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::size"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::type"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::type"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10kBYTE_TYPEE", "tensorrt_llm::runtime::BufferManager::kBYTE_TYPE"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager5mPoolE", "tensorrt_llm::runtime::BufferManager::mPool"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7mStreamE", "tensorrt_llm::runtime::BufferManager::mStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager9mTrimPoolE", "tensorrt_llm::runtime::BufferManager::mTrimPool"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolFreeEv", "tensorrt_llm::runtime::BufferManager::memoryPoolFree"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager18memoryPoolReservedEv", "tensorrt_llm::runtime::BufferManager::memoryPoolReserved"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", "tensorrt_llm::runtime::BufferManager::memoryPoolTrimTo"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", "tensorrt_llm::runtime::BufferManager::memoryPoolTrimTo::size"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolUsedEv", "tensorrt_llm::runtime::BufferManager::memoryPoolUsed"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem::buffer"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem::value"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", "tensorrt_llm::runtime::BufferManager::setZero"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", "tensorrt_llm::runtime::BufferManager::setZero::buffer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerD0Ev", "tensorrt_llm::runtime::BufferManager::~BufferManager"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", "tensorrt_llm::runtime::BufferRange"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange4BaseE", "tensorrt_llm::runtime::BufferRange::Base"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::U"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::U"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange::size"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", "tensorrt_llm::runtime::BufferRange::T"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEventE", "tensorrt_llm::runtime::CudaEvent"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", "tensorrt_llm::runtime::CudaEvent::CudaEvent"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent::event"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", "tensorrt_llm::runtime::CudaEvent::CudaEvent::flags"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent::ownsEvent"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7DeleterE", "tensorrt_llm::runtime::CudaEvent::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEv", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter::ownsEvent"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter10mOwnsEventE", "tensorrt_llm::runtime::CudaEvent::Deleter::mOwnsEvent"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", "tensorrt_llm::runtime::CudaEvent::Deleter::operator()"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", "tensorrt_llm::runtime::CudaEvent::Deleter::operator()::event"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent8EventPtrE", "tensorrt_llm::runtime::CudaEvent::EventPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent12element_typeE", "tensorrt_llm::runtime::CudaEvent::element_type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent3getEv", "tensorrt_llm::runtime::CudaEvent::get"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent6mEventE", "tensorrt_llm::runtime::CudaEvent::mEvent"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaEvent::pointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent11synchronizeEv", "tensorrt_llm::runtime::CudaEvent::synchronize"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStreamE", "tensorrt_llm::runtime::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::device"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream::flags"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::ownsStream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream::priority"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::CudaStream::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::stream"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7DeleterE", "tensorrt_llm::runtime::CudaStream::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEv", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter::ownsStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter11mOwnsStreamE", "tensorrt_llm::runtime::CudaStream::Deleter::mOwnsStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::Deleter::operator()"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::Deleter::operator()::stream"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream9StreamPtrE", "tensorrt_llm::runtime::CudaStream::StreamPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream3getEv", "tensorrt_llm::runtime::CudaStream::get"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream9getDeviceEv", "tensorrt_llm::runtime::CudaStream::getDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mDeviceE", "tensorrt_llm::runtime::CudaStream::mDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mStreamE", "tensorrt_llm::runtime::CudaStream::mStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::record"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::record"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::record::event"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::record::event"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream11synchronizeEv", "tensorrt_llm::runtime::CudaStream::synchronize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::wait"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::wait"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::wait::event"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::wait::event"], [1, 2, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kDataType"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kIsPointer"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kIsUnsigned"], [1, 2, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::kDataType"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::type"], [1, 2, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>"], [1, 8, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::type"], [1, 2, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>"], [1, 8, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::type"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInputE", "tensorrt_llm::runtime::DecodingInput"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::endIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::maxLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::sinkTokenLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedPathIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::chunkedContextNextTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::seqSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs12acceptedLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15acceptedPathIdsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedPathIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14acceptedTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::chunkedContextNextTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13lastDraftLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14lastDraftPathsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15lastDraftTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13nextDraftLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14nextDraftPathsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs8seqSlotsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::seqSlots"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::bestPathIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::bestPathLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16lastDraftIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15lastDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21lastGenerationLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs19lastPositionIdsBaseE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastPositionIdsBase"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs5masksE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::masks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs18maxGenLengthDeviceE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::maxGenLengthDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16nextDraftIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextDraftProbsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextFlatTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextFlatTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21nextGenerationLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs17packedPositionIdsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::packedPositionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs8seqSlotsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::seqSlots"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs17constantThresholdE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::constantThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11draftLogitsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs10draftProbsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs13draftTokenIdsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftTokenIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14numDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::numDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18numDraftTokensHostE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::numDraftTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs4stepE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::step"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11targetProbsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::targetProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14useDraftLogitsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useDraftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18useDraftLogitsHostE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useDraftLogitsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs28useRandomAcceptanceThresholdE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useRandomAcceptanceThreshold"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputsE", "tensorrt_llm::runtime::DecodingInput::LookaheadInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputs13tokensPerStepE", "tensorrt_llm::runtime::DecodingInput::LookaheadInputs::tokensPerStep"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs22medusaCurTokensPerStepE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaCurTokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs12medusaLogitsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs11medusaPathsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs25medusaTargetTokensPerStepE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaTargetTokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs13medusaTreeIdsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaTreeIds"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14TensorConstPtrE", "tensorrt_llm::runtime::DecodingInput::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9TensorPtrE", "tensorrt_llm::runtime::DecodingInput::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsLensE", "tensorrt_llm::runtime::DecodingInput::badWordsLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13badWordsListsE", "tensorrt_llm::runtime::DecodingInput::badWordsLists"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsPtrsE", "tensorrt_llm::runtime::DecodingInput::badWordsPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9batchSizeE", "tensorrt_llm::runtime::DecodingInput::batchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10batchSlotsE", "tensorrt_llm::runtime::DecodingInput::batchSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10beamWidthsE", "tensorrt_llm::runtime::DecodingInput::beamWidths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput16cacheIndirectionE", "tensorrt_llm::runtime::DecodingInput::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11eagleInputsE", "tensorrt_llm::runtime::DecodingInput::eagleInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13embeddingBiasE", "tensorrt_llm::runtime::DecodingInput::embeddingBias"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6endIdsE", "tensorrt_llm::runtime::DecodingInput::endIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25explicitDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::explicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25externalDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::externalDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13finishReasonsE", "tensorrt_llm::runtime::DecodingInput::finishReasons"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15generationStepsE", "tensorrt_llm::runtime::DecodingInput::generationSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput7lengthsE", "tensorrt_llm::runtime::DecodingInput::lengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6logitsE", "tensorrt_llm::runtime::DecodingInput::logits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9logitsVecE", "tensorrt_llm::runtime::DecodingInput::logitsVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15lookaheadInputsE", "tensorrt_llm::runtime::DecodingInput::lookaheadInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput18maxAttentionWindowE", "tensorrt_llm::runtime::DecodingInput::maxAttentionWindow"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14maxBadWordsLenE", "tensorrt_llm::runtime::DecodingInput::maxBadWordsLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9maxLengthE", "tensorrt_llm::runtime::DecodingInput::maxLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15maxStopWordsLenE", "tensorrt_llm::runtime::DecodingInput::maxStopWordsLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12medusaInputsE", "tensorrt_llm::runtime::DecodingInput::medusaInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput17noRepeatNgramSizeE", "tensorrt_llm::runtime::DecodingInput::noRepeatNgramSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput19sequenceLimitLengthE", "tensorrt_llm::runtime::DecodingInput::sequenceLimitLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15sinkTokenLengthE", "tensorrt_llm::runtime::DecodingInput::sinkTokenLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput4stepE", "tensorrt_llm::runtime::DecodingInput::step"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsLensE", "tensorrt_llm::runtime::DecodingInput::stopWordsLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14stopWordsListsE", "tensorrt_llm::runtime::DecodingInput::stopWordsLists"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsPtrsE", "tensorrt_llm::runtime::DecodingInput::stopWordsPtrs"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutputE", "tensorrt_llm::runtime::DecodingOutput"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypothesesE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses10batchDonesE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::batchDones"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses14cumLogProbsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::cumLogProbsCBA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::empty"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::empty::manager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init::endId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init::manager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11logProbsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::logProbsCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18minNormedScoresCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::minNormedScoresCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses15normedScoresCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::normedScoresCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11numBeamsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::numBeamsCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses12outputIdsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::outputIdsCBA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7releaseEv", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::release"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::beamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::maxSequenceLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18sequenceLengthsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::sequenceLengthsCBA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice::batchIndex"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput::gatheredIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput::ids"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputsE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs21acceptedLengthsCumSumE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::acceptedLengthsCumSum"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs17acceptedTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::acceptedTokensLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18nextDraftTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::nextDraftTokensLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs12pathsOffsetsE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::pathsOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18prevDraftTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::prevDraftTokensLen"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9TensorPtrE", "tensorrt_llm::runtime::DecodingOutput::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14beamHypothesesE", "tensorrt_llm::runtime::DecodingOutput::beamHypotheses"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16cacheIndirectionE", "tensorrt_llm::runtime::DecodingOutput::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11cumLogProbsE", "tensorrt_llm::runtime::DecodingOutput::cumLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12eagleBuffersE", "tensorrt_llm::runtime::DecodingOutput::eagleBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26explicitDraftTokensBuffersE", "tensorrt_llm::runtime::DecodingOutput::explicitDraftTokensBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13finishReasonsE", "tensorrt_llm::runtime::DecodingOutput::finishReasons"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11finishedSumE", "tensorrt_llm::runtime::DecodingOutput::finishedSum"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11gatheredIdsE", "tensorrt_llm::runtime::DecodingOutput::gatheredIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput3idsE", "tensorrt_llm::runtime::DecodingOutput::ids"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput17kNegativeInfinityE", "tensorrt_llm::runtime::DecodingOutput::kNegativeInfinity"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput7lengthsE", "tensorrt_llm::runtime::DecodingOutput::lengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput8logProbsE", "tensorrt_llm::runtime::DecodingOutput::logProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13logProbsTiledE", "tensorrt_llm::runtime::DecodingOutput::logProbsTiled"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16lookaheadOutputsE", "tensorrt_llm::runtime::DecodingOutput::lookaheadOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9newTokensE", "tensorrt_llm::runtime::DecodingOutput::newTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14newTokensStepsE", "tensorrt_llm::runtime::DecodingOutput::newTokensSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12newTokensVecE", "tensorrt_llm::runtime::DecodingOutput::newTokensVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9parentIdsE", "tensorrt_llm::runtime::DecodingOutput::parentIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26speculativeDecodingOutputsE", "tensorrt_llm::runtime::DecodingOutput::speculativeDecodingOutputs"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", "tensorrt_llm::runtime::DeviceAllocationNvls"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls20DeviceAllocationNvlsEv", "tensorrt_llm::runtime::DeviceAllocationNvls::DeviceAllocationNvls"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", "tensorrt_llm::runtime::DeviceAllocationNvls::T"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls9_capacityE", "tensorrt_llm::runtime::DeviceAllocationNvls::_capacity"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls7_handleE", "tensorrt_llm::runtime::DeviceAllocationNvls::_handle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls4freeEv", "tensorrt_llm::runtime::DeviceAllocationNvls::free"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls11getCapacityEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getCapacity"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls21getIpcUnicastPointersEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getIpcUnicastPointers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls19getMulticastPointerEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getMulticastPointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls17getUnicastPointerEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getUnicastPointer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvlsD0Ev", "tensorrt_llm::runtime::DeviceAllocationNvls::~DeviceAllocationNvls"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffersE", "tensorrt_llm::runtime::EagleBuffers"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9BufferPtrE", "tensorrt_llm::runtime::EagleBuffers::BufferPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::decodingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs12acceptedLensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13acceptedPathsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14acceptedTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::chunkedContextNextTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13nextDraftLensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14nextDraftPathsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs15nextDraftTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7ITensorE", "tensorrt_llm::runtime::EagleBuffers::ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6InputsE", "tensorrt_llm::runtime::EagleBuffers::Inputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22allLayersDraftTokenIdsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersDraftTokenIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33allLayersDraftTokenIdsPredecessorE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersDraftTokenIdsPredecessor"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15allLayersScoresE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersScores"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::chunkedContextNextTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20currentExpandIndicesE", "tensorrt_llm::runtime::EagleBuffers::Inputs::currentExpandIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs9draftLensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10draftPathsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14draftPathsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftPathsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs11draftTokensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22dynamicTreeMaxTopKHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::dynamicTreeMaxTopKHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetCtxContextLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxContextLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetCtxPastKeyValueLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxPastKeyValueLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetCtxRequestTypesHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxRequestTypesHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetGenContextLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenContextLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetGenPastKeyValueLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenPastKeyValueLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetGenRequestTypesHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenRequestTypesHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18inputGenTokensHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::inputGenTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14posteriorAlphaE", "tensorrt_llm::runtime::EagleBuffers::Inputs::posteriorAlpha"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18posteriorThresholdE", "tensorrt_llm::runtime::EagleBuffers::Inputs::posteriorThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10prevScoresE", "tensorrt_llm::runtime::EagleBuffers::Inputs::prevScores"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs16randomDataSampleE", "tensorrt_llm::runtime::EagleBuffers::Inputs::randomDataSample"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20randomDataValidationE", "tensorrt_llm::runtime::EagleBuffers::Inputs::randomDataValidation"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29specDecodingGenerationLengthsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33specDecodingGenerationLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingGenerationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs23specDecodingPackedMasksE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27specDecodingPositionOffsetsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingPositionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs12temperaturesE", "tensorrt_llm::runtime::EagleBuffers::Inputs::temperatures"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18useDynamicTreeHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::useDynamicTreeHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15useSpecDecodingE", "tensorrt_llm::runtime::EagleBuffers::Inputs::useSpecDecoding"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13LlmRequestPtrE", "tensorrt_llm::runtime::EagleBuffers::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13RequestVectorE", "tensorrt_llm::runtime::EagleBuffers::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers10SizeType32E", "tensorrt_llm::runtime::EagleBuffers::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorMapE", "tensorrt_llm::runtime::EagleBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorPtrE", "tensorrt_llm::runtime::EagleBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers28chunkedContextNextTokensHostE", "tensorrt_llm::runtime::EagleBuffers::chunkedContextNextTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers23cumSumGenerationLengthsE", "tensorrt_llm::runtime::EagleBuffers::cumSumGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12engineInputsE", "tensorrt_llm::runtime::EagleBuffers::engineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13engineOutputsE", "tensorrt_llm::runtime::EagleBuffers::engineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18greedySamplingHostE", "tensorrt_llm::runtime::EagleBuffers::greedySamplingHost"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26mDefaultPosteriorThresholdE", "tensorrt_llm::runtime::EagleBuffers::mDefaultPosteriorThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers17mDoGreedySamplingE", "tensorrt_llm::runtime::EagleBuffers::mDoGreedySampling"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers19maxGenerationLengthE", "tensorrt_llm::runtime::EagleBuffers::maxGenerationLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18posteriorAlphaHostE", "tensorrt_llm::runtime::EagleBuffers::posteriorAlphaHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers22posteriorThresholdHostE", "tensorrt_llm::runtime::EagleBuffers::posteriorThresholdHost"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::numGenSequences"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers21scanReduceTempStorageE", "tensorrt_llm::runtime::EagleBuffers::scanReduceTempStorage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26scanReduceTempStorageBytesE", "tensorrt_llm::runtime::EagleBuffers::scanReduceTempStorageBytes"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::contextRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::contextRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::decoderBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::draftBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::eagleModule"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::genRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::genRequests"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::vocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModuleE", "tensorrt_llm::runtime::EagleModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleEv", "tensorrt_llm::runtime::EagleModule::EagleModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxNonLeafNodesPerLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::numTransformersLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule22getDefaultEagleChoicesEv", "tensorrt_llm::runtime::EagleModule::getDefaultEagleChoices"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule26getMaxNonLeafNodesPerLayerEv", "tensorrt_llm::runtime::EagleModule::getMaxNonLeafNodesPerLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule23getNumTransformerLayersEv", "tensorrt_llm::runtime::EagleModule::getNumTransformerLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule20mDefaultEagleChoicesE", "tensorrt_llm::runtime::EagleModule::mDefaultEagleChoices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule24mMaxNonLeafNodesPerLayerE", "tensorrt_llm::runtime::EagleModule::mMaxNonLeafNodesPerLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule21mNumTransformersLayerE", "tensorrt_llm::runtime::EagleModule::mNumTransformersLayer"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffersE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9BufferPtrE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::BufferPtr"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs15positionOffsetsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs::positionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs18requestTypesDeviceE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs::requestTypesDevice"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::bestPathIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::bestPathLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs5masksE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::masks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs11maxGenTokenE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::maxGenToken"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs16nextDraftIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextDraftProbsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15nextDraftTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextFlatTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextFlatTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs21nextGenerationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs19nextPositionOffsetsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextPositionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs17packedPositionIdsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::packedPositionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs13totalGenTokenE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::totalGenToken"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7ITensorE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12draftIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs10draftProbsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11draftTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs17generationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::generationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs21generationLengthsHostE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::generationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16maxGenLengthHostE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::maxGenLengthHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11packedMasksE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::packedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11positionIdsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::positionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15positionIdsBaseE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::positionIdsBase"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16randomDataSampleE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::randomDataSample"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs20randomDataValidationE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::randomDataValidation"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12temperaturesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::temperatures"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15useSpecDecodingE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::useSpecDecoding"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers10SizeType32E", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorMapE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorPtrE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers23cumSumGenerationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::cumSumGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12engineInputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::engineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13engineOutputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::engineOutputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::numGenSequences"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers15scanTempStorageE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::scanTempStorage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers20scanTempStorageBytesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::scanTempStorageBytes"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::contextPositionIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::contextPositionIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::decoderBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::draftBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::explicitDraftTokensModule"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::stream"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::stream"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::vocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::worldConfig"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", "tensorrt_llm::runtime::GenericPromptTuningParams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::embeddingTable"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::tasks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::vocabSize"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams10SizeType32E", "tensorrt_llm::runtime::GenericPromptTuningParams::SizeType32"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", "tensorrt_llm::runtime::GenericPromptTuningParams::TTensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9TensorPtrE", "tensorrt_llm::runtime::GenericPromptTuningParams::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams14embeddingTableE", "tensorrt_llm::runtime::GenericPromptTuningParams::embeddingTable"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams19promptTuningEnabledE", "tensorrt_llm::runtime::GenericPromptTuningParams::promptTuningEnabled"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams5tasksE", "tensorrt_llm::runtime::GenericPromptTuningParams::tasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9vocabSizeE", "tensorrt_llm::runtime::GenericPromptTuningParams::vocabSize"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", "tensorrt_llm::runtime::GptDecoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13CudaStreamPtrE", "tensorrt_llm::runtime::GptDecoder::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::speculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::vocabSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::vocabSizePadded"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", "tensorrt_llm::runtime::GptDecoder::T"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder9TensorPtrE", "tensorrt_llm::runtime::GptDecoder::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder17getSamplingConfigEv", "tensorrt_llm::runtime::GptDecoder::getSamplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder23mDecodingLayerWorkspaceE", "tensorrt_llm::runtime::GptDecoder::mDecodingLayerWorkspace"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mDecodingModeE", "tensorrt_llm::runtime::GptDecoder::mDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder19mDynamicDecodeLayerE", "tensorrt_llm::runtime::GptDecoder::mDynamicDecodeLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder8mManagerE", "tensorrt_llm::runtime::GptDecoder::mManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mMaxBatchSizeE", "tensorrt_llm::runtime::GptDecoder::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder15mSamplingConfigE", "tensorrt_llm::runtime::GptDecoder::mSamplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10mVocabSizeE", "tensorrt_llm::runtime::GptDecoder::mVocabSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16mVocabSizePaddedE", "tensorrt_llm::runtime::GptDecoder::mVocabSizePadded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::requests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::samplingConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatchedE", "tensorrt_llm::runtime::GptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13CudaStreamPtrE", "tensorrt_llm::runtime::GptDecoderBatched::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::speculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::stream"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13GptDecoderPtrE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13LlmRequestPtrE", "tensorrt_llm::runtime::GptDecoderBatched::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13RequestVectorE", "tensorrt_llm::runtime::GptDecoderBatched::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14SharedConstPtrE", "tensorrt_llm::runtime::GptDecoderBatched::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched9TensorPtrE", "tensorrt_llm::runtime::GptDecoderBatched::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::batchSlot"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::decoderState"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::samplingConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::streaming"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch::output"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getBufferManagerEv", "tensorrt_llm::runtime::GptDecoderBatched::getBufferManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getDecoderStreamEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched20getUnderlyingDecoderEv", "tensorrt_llm::runtime::GptDecoderBatched::getUnderlyingDecoder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mBufferManagerE", "tensorrt_llm::runtime::GptDecoderBatched::mBufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched8mDecoderE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13mDecoderStateE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoderState"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mDecoderStreamE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoderStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mRuntimeStreamE", "tensorrt_llm::runtime::GptDecoderBatched::mRuntimeStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::step"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setEagleInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setEagleInputs::input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setExplicitDraftTokensInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setExplicitDraftTokensInputs::input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxTokensPerStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfigE", "tensorrt_llm::runtime::GptJsonConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::name"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::precision"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::runtimeDefaults"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::tensorParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::version"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", "tensorrt_llm::runtime::GptJsonConfig::engineFilename"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::model"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::worldConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig21getContextParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getContextParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getGpusPerNodeEv", "tensorrt_llm::runtime::GptJsonConfig::getGpusPerNode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getModelConfigEv", "tensorrt_llm::runtime::GptJsonConfig::getModelConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig21getModelConfigMutableEv", "tensorrt_llm::runtime::GptJsonConfig::getModelConfigMutable"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig7getNameEv", "tensorrt_llm::runtime::GptJsonConfig::getName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig22getPipelineParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getPipelineParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getPrecisionEv", "tensorrt_llm::runtime::GptJsonConfig::getPrecision"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig18getRuntimeDefaultsEv", "tensorrt_llm::runtime::GptJsonConfig::getRuntimeDefaults"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig20getTensorParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getTensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig10getVersionEv", "tensorrt_llm::runtime::GptJsonConfig::getVersion"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getWorldSizeEv", "tensorrt_llm::runtime::GptJsonConfig::getWorldSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig19mContextParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mContextParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mGpusPerNodeE", "tensorrt_llm::runtime::GptJsonConfig::mGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mModelConfigE", "tensorrt_llm::runtime::GptJsonConfig::mModelConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5mNameE", "tensorrt_llm::runtime::GptJsonConfig::mName"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig20mPipelineParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mPipelineParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig10mPrecisionE", "tensorrt_llm::runtime::GptJsonConfig::mPrecision"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig16mRuntimeDefaultsE", "tensorrt_llm::runtime::GptJsonConfig::mRuntimeDefaults"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig18mTensorParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mTensorParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig8mVersionE", "tensorrt_llm::runtime::GptJsonConfig::mVersion"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::parse::json"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", "tensorrt_llm::runtime::GptJsonConfig::parse::json"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", "tensorrt_llm::runtime::GptJsonConfig::parse::path"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferE", "tensorrt_llm::runtime::IBuffer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer8DataTypeE", "tensorrt_llm::runtime::IBuffer::DataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferERK7IBuffer", "tensorrt_llm::runtime::IBuffer::IBuffer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferEv", "tensorrt_llm::runtime::IBuffer::IBuffer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::IBuffer::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9SharedPtrE", "tensorrt_llm::runtime::IBuffer::SharedPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14UniqueConstPtrE", "tensorrt_llm::runtime::IBuffer::UniqueConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9UniquePtrE", "tensorrt_llm::runtime::IBuffer::UniquePtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataEv", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataEv", "tensorrt_llm::runtime::IBuffer::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data::index"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data::index"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getCapacityEv", "tensorrt_llm::runtime::IBuffer::getCapacity"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getDataTypeEv", "tensorrt_llm::runtime::IBuffer::getDataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", "tensorrt_llm::runtime::IBuffer::getDataTypeName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer15getDataTypeNameEv", "tensorrt_llm::runtime::IBuffer::getDataTypeName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", "tensorrt_llm::runtime::IBuffer::getDataTypeName::dataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer13getMemoryTypeEv", "tensorrt_llm::runtime::IBuffer::getMemoryType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer17getMemoryTypeNameEv", "tensorrt_llm::runtime::IBuffer::getMemoryTypeName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7getSizeEv", "tensorrt_llm::runtime::IBuffer::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer14getSizeInBytesEv", "tensorrt_llm::runtime::IBuffer::getSizeInBytes"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", "tensorrt_llm::runtime::IBuffer::memoryType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", "tensorrt_llm::runtime::IBuffer::memoryType::data"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferaSERK7IBuffer", "tensorrt_llm::runtime::IBuffer::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7releaseEv", "tensorrt_llm::runtime::IBuffer::release"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", "tensorrt_llm::runtime::IBuffer::resize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", "tensorrt_llm::runtime::IBuffer::resize::newSize"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::TConstPtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::tensor"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", "tensorrt_llm::runtime::IBuffer::toBytes"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", "tensorrt_llm::runtime::IBuffer::toBytes::size"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", "tensorrt_llm::runtime::IBuffer::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::capacity"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::capacity"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::type"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferD0Ev", "tensorrt_llm::runtime::IBuffer::~IBuffer"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderE", "tensorrt_llm::runtime::IGptDecoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder14TensorConstPtrE", "tensorrt_llm::runtime::IGptDecoder::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder9TensorPtrE", "tensorrt_llm::runtime::IGptDecoder::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::speculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::vocabSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::vocabSizePadded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder17getSamplingConfigEv", "tensorrt_llm::runtime::IGptDecoder::getSamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::requests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderD0Ev", "tensorrt_llm::runtime::IGptDecoder::~IGptDecoder"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedE", "tensorrt_llm::runtime::IGptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13CudaStreamPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched18IGptDecoderBatchedEv", "tensorrt_llm::runtime::IGptDecoderBatched::IGptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13LlmRequestPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13RequestVectorE", "tensorrt_llm::runtime::IGptDecoderBatched::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched9TensorPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::batchSlot"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::decoderState"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::samplingConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::streaming"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxTokensPerStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedD0Ev", "tensorrt_llm::runtime::IGptDecoderBatched::~IGptDecoderBatched"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7ITensorE", "tensorrt_llm::runtime::ITensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9DimType64E", "tensorrt_llm::runtime::ITensor::DimType64"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorERK7ITensor", "tensorrt_llm::runtime::ITensor::ITensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorEv", "tensorrt_llm::runtime::ITensor::ITensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5ShapeE", "tensorrt_llm::runtime::ITensor::Shape"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::ITensor::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9SharedPtrE", "tensorrt_llm::runtime::ITensor::SharedPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9TensorMapE", "tensorrt_llm::runtime::ITensor::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor14UniqueConstPtrE", "tensorrt_llm::runtime::ITensor::UniqueConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9UniquePtrE", "tensorrt_llm::runtime::ITensor::UniquePtr"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", "tensorrt_llm::runtime::ITensor::castSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", "tensorrt_llm::runtime::ITensor::castSize::newSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN::sliceN"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN::tensor"], [1, 3, 1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", "tensorrt_llm::runtime::ITensor::getDimension"], [1, 8, 1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", "tensorrt_llm::runtime::ITensor::getDimension::n"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor8getShapeEv", "tensorrt_llm::runtime::ITensor::getShape"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::makeShape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::makeShape::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensoraSERK7ITensor", "tensorrt_llm::runtime::ITensor::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", "tensorrt_llm::runtime::ITensor::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", "tensorrt_llm::runtime::ITensor::reshape::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", "tensorrt_llm::runtime::ITensor::resize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", "tensorrt_llm::runtime::ITensor::resize::newSize"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::count"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::count"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::dims"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::lhs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::lhs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::other"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", "tensorrt_llm::runtime::ITensor::shapeEquals::other"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::rhs"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::shape"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", "tensorrt_llm::runtime::ITensor::strides"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", "tensorrt_llm::runtime::ITensor::strides::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", "tensorrt_llm::runtime::ITensor::toString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", "tensorrt_llm::runtime::ITensor::toString::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::shape"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", "tensorrt_llm::runtime::ITensor::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::TConstPtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view::dims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", "tensorrt_llm::runtime::ITensor::view::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", "tensorrt_llm::runtime::ITensor::volume"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", "tensorrt_llm::runtime::ITensor::volume::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", "tensorrt_llm::runtime::ITensor::volumeNonNegative"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", "tensorrt_llm::runtime::ITensor::volumeNonNegative::shape"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::capacity"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::capacity"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::type"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensorD0Ev", "tensorrt_llm::runtime::ITensor::~ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryE", "tensorrt_llm::runtime::IpcMemory"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9BufferPtrE", "tensorrt_llm::runtime::IpcMemory::BufferPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory10FLAGS_SIZEE", "tensorrt_llm::runtime::IpcMemory::FLAGS_SIZE"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERK9IpcMemory", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERR9IpcMemory", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::bufferSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::openIpc"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::bufferSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory16destroyIpcMemoryEv", "tensorrt_llm::runtime::IpcMemory::destroyIpcMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9IpcMemory11getCommPtrsEv", "tensorrt_llm::runtime::IpcMemory::getCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mBufferE", "tensorrt_llm::runtime::IpcMemory::mBuffer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9mCommPtrsE", "tensorrt_llm::runtime::IpcMemory::mCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory8mOpenIpcE", "tensorrt_llm::runtime::IpcMemory::mOpenIpc"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mTpRankE", "tensorrt_llm::runtime::IpcMemory::mTpRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERK9IpcMemory", "tensorrt_llm::runtime::IpcMemory::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERR9IpcMemory", "tensorrt_llm::runtime::IpcMemory::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryD0Ev", "tensorrt_llm::runtime::IpcMemory::~IpcMemory"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandleE", "tensorrt_llm::runtime::IpcNvlsHandle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle14ipc_uc_handlesE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_handles"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle11ipc_uc_ptrsE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_ptrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle10ipc_uc_vasE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_vas"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9mc_handleE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_handle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6mc_ptrE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_ptr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5mc_vaE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_va"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle4sizeE", "tensorrt_llm::runtime::IpcNvlsHandle::size"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9uc_handleE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_handle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6uc_ptrE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_ptr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5uc_vaE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_va"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffersE", "tensorrt_llm::runtime::LookaheadDecodingBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::maxTokensPerStep"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers9TensorPtrE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers17generationLengthsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::generationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11packedMasksE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::packedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11positionIdsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::positionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers15positionOffsetsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::positionOffsets"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModuleE", "tensorrt_llm::runtime::LookaheadModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleEv", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule::maxDraftPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime15LookaheadModule18getExecutionConfigEv", "tensorrt_llm::runtime::LookaheadModule::getExecutionConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule16mExecutionConfigE", "tensorrt_llm::runtime::LookaheadModule::mExecutionConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", "tensorrt_llm::runtime::LookaheadModule::setExecutionConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", "tensorrt_llm::runtime::LookaheadModule::setExecutionConfig::config"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffersE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::decodingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::runtime"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorMapE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorPtrE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18batchSlotsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::batchSlotsHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers12cumSumLengthE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::cumSumLength"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers24disableLookaheadDecodingEv", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::disableLookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding::tokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23generationLengthsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21generationLengthsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers25generationLengthsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsHostCopy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers14packedMaskHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMaskHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18packedMaskHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMaskHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17packedMasksDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMasksDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17positionIdsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15positionIdsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionIdsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21positionOffsetsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionOffsetsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23positionOffsetsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsHostCopy"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::numGenSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::tokensPerStep"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::decoderLookaheadBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::runtime"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15useSpecDecodingE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::useSpecDecoding"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCacheE", "tensorrt_llm::runtime::LoraCache"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::pageManagerConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10TaskIdTypeE", "tensorrt_llm::runtime::LoraCache::TaskIdType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig11adapterSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::adapterSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6inSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::inSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7layerIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::layerId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8moduleIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::moduleId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8numSlotsE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::numSlots"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::operator==::o"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7outSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::outSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6pageIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::pageId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17scalingVecPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::scalingVecPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7slotIdxE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::slotIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8toStringEv", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::toString"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig16weightsInPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::weightsInPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17weightsOutPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::weightsOutPointer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache28TaskLayerModuleConfigListPtrE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfigListPtr"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueE", "tensorrt_llm::runtime::LoraCache::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueEv", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::configs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::done"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::inProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::it"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::loadInProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::loaded"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::o"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::pageIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7configsE", "tensorrt_llm::runtime::LoraCache::TaskValue::configs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue4doneE", "tensorrt_llm::runtime::LoraCache::TaskValue::done"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue10inProgressE", "tensorrt_llm::runtime::LoraCache::TaskValue::inProgress"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue2itE", "tensorrt_llm::runtime::LoraCache::TaskValue::it"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue14loadInProgressE", "tensorrt_llm::runtime::LoraCache::TaskValue::loadInProgress"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue6loadedE", "tensorrt_llm::runtime::LoraCache::TaskValue::loaded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::operator=::o"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7pageIdsE", "tensorrt_llm::runtime::LoraCache::TaskValue::pageIds"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueD0Ev", "tensorrt_llm::runtime::LoraCache::TaskValue::~TaskValue"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12TaskValuePtrE", "tensorrt_llm::runtime::LoraCache::TaskValuePtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TensorPtrE", "tensorrt_llm::runtime::LoraCache::TensorPtr"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatusE", "tensorrt_llm::runtime::LoraCache::ValueStatus"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_LOADED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_MISSING"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_PROCESSING"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bump"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bump::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bumpTaskInProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bumpTaskInProgress::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", "tensorrt_llm::runtime::LoraCache::claimPagesWithEvict"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", "tensorrt_llm::runtime::LoraCache::claimPagesWithEvict::numPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::deviceCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::markDone"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::sourceTaskValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetPageIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetTaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::moduleIdToModel"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::pageIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::pages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::weights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", "tensorrt_llm::runtime::LoraCache::determineNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", "tensorrt_llm::runtime::LoraCache::determineNumPages"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", "tensorrt_llm::runtime::LoraCache::determineNumPages::config"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", "tensorrt_llm::runtime::LoraCache::determineNumPages::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", "tensorrt_llm::runtime::LoraCache::fits"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", "tensorrt_llm::runtime::LoraCache::fits::config"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", "tensorrt_llm::runtime::LoraCache::get"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", "tensorrt_llm::runtime::LoraCache::get::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache11getNumPagesEv", "tensorrt_llm::runtime::LoraCache::getNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", "tensorrt_llm::runtime::LoraCache::getPagePtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", "tensorrt_llm::runtime::LoraCache::getPagePtr::pageId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", "tensorrt_llm::runtime::LoraCache::getStatus"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", "tensorrt_llm::runtime::LoraCache::getStatus::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", "tensorrt_llm::runtime::LoraCache::has"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", "tensorrt_llm::runtime::LoraCache::has::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isDone"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isDone::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isLoaded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isLoaded::taskId"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_LOADED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_MISSING"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_PROCESSING"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::cacheValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::taskId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::weights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::weights"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache14mBufferManagerE", "tensorrt_llm::runtime::LoraCache::mBufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9mCacheMapE", "tensorrt_llm::runtime::LoraCache::mCacheMap"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mCacheMutexE", "tensorrt_llm::runtime::LoraCache::mCacheMutex"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mCachePageManagerE", "tensorrt_llm::runtime::LoraCache::mCachePageManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21mDeviceBufferManagersE", "tensorrt_llm::runtime::LoraCache::mDeviceBufferManagers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10mDoneTasksE", "tensorrt_llm::runtime::LoraCache::mDoneTasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16mInProgressTasksE", "tensorrt_llm::runtime::LoraCache::mInProgressTasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mModelConfigE", "tensorrt_llm::runtime::LoraCache::mModelConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mModuleIdToModuleE", "tensorrt_llm::runtime::LoraCache::mModuleIdToModule"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18mPageManagerConfigE", "tensorrt_llm::runtime::LoraCache::mPageManagerConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mPagesMutexE", "tensorrt_llm::runtime::LoraCache::mPagesMutex"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mWorldConfigE", "tensorrt_llm::runtime::LoraCache::mWorldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11markAllDoneEv", "tensorrt_llm::runtime::LoraCache::markAllDone"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::markTaskDone"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::markTaskDone::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::load"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::taskId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::weights"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::tpRank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::tpSize"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::input"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::output"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::tpRank"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::tpSize"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionE", "tensorrt_llm::runtime::LoraCacheFullException"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraCacheFullException::LoraCacheFullException"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraCacheFullException::LoraCacheFullException::msg"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionD0Ev", "tensorrt_llm::runtime::LoraCacheFullException::~LoraCacheFullException"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManagerE", "tensorrt_llm::runtime::LoraCachePageManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager::config"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager9TensorPtrE", "tensorrt_llm::runtime::LoraCachePageManager::TensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::blockPtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::blockPtr::blockIdx"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::claimPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::claimPages::numPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::initialize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::initialize::bufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager7mConfigE", "tensorrt_llm::runtime::LoraCachePageManager::mConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12mFreePageIdsE", "tensorrt_llm::runtime::LoraCachePageManager::mFreePageIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mIsPageFreeE", "tensorrt_llm::runtime::LoraCachePageManager::mIsPageFree"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mPageBlocksE", "tensorrt_llm::runtime::LoraCachePageManager::mPageBlocks"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::mutablePagePtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::mutablePagePtr::pageIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager17numAvailablePagesEv", "tensorrt_llm::runtime::LoraCachePageManager::numAvailablePages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::pagePtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::pagePtr::pageIdx"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCachePageManager::releasePages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCachePageManager::releasePages::pages"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfigE", "tensorrt_llm::runtime::LoraCachePageManagerConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::dType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::maxPagesPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::memType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::numCopyStreams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::pageWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::slotsPerPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::totalNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig11getDataTypeEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getInitToZeroEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getInitToZero"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig19getMaxPagesPerBlockEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getMaxPagesPerBlock"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getMemoryTypeEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getMemoryType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig17getNumCopyStreamsEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getNumCopyStreams"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig12getPageWidthEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getPageWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig15getSlotsPerPageEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getSlotsPerPage"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig16getTotalNumPagesEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getTotalNumPages"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig9mDataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mInitToZeroE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mInitToZero"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17mMaxPagesPerBlockE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mMaxPagesPerBlock"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mMemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mMemoryType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15mNumCopyStreamsE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mNumCopyStreams"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig10mPageWidthE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mPageWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13mSlotsPerPageE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mSlotsPerPage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig14mTotalNumPagesE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mTotalNumPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setDataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setDataType::dtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setInitToZero"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setInitToZero::initToZero"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMaxPagesPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMaxPagesPerBlock::maxPagesPerBlock"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMemoryType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMemoryType::memoryType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setNumCopyStreams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setNumCopyStreams::numCopyStreams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setPageWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setPageWidth::pageWidth"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setSlotsPerPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setSlotsPerPage::slotsPerPage"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setTotalNumPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setTotalNumPage::totalNumPages"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionE", "tensorrt_llm::runtime::LoraExpectedException"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraExpectedException::LoraExpectedException"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraExpectedException::LoraExpectedException::msg"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionD0Ev", "tensorrt_llm::runtime::LoraExpectedException::~LoraExpectedException"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleE", "tensorrt_llm::runtime::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleEv", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inDimFirst"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inTpSplitDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", "tensorrt_llm::runtime::LoraModule::LoraModule::o"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outDimFirst"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outTpSplitDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::t"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleTypeE", "tensorrt_llm::runtime::LoraModule::ModuleType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kATTN_DENSEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_DENSE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_KE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_K"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_QE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_Q"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kATTN_QKVE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_QKV"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_VE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_V"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType17kCROSS_ATTN_DENSEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_DENSE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_KE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_K"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_QE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_Q"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType15kCROSS_ATTN_QKVE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_QKV"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_VE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_V"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType8kINVALIDE", "tensorrt_llm::runtime::LoraModule::ModuleType::kINVALID"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_4H_TO_HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_4H_TO_H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMLP_GATEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_GATE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_GATE_UPE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_GATE_UP"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_H_TO_4HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_H_TO_4H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMLP_ROUTERE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_ROUTER"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_4H_TO_HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_4H_TO_H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMOE_GATEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_GATE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_H_TO_4HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_H_TO_4H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMOE_ROUTERE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_ROUTER"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule9TensorPtrE", "tensorrt_llm::runtime::LoraModule::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::attentionHeadSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::loraModuleNames"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::mlpHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numAttentionHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numExperts"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numKvAttentionHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize::isDora"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5inDimEv", "tensorrt_llm::runtime::LoraModule::inDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10inDimFirstEv", "tensorrt_llm::runtime::LoraModule::inDimFirst"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::inSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::inSize::adapterSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12inTpSplitDimEv", "tensorrt_llm::runtime::LoraModule::inTpSplitDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localInDim"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localInDim::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localOutDim"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localOutDim::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize::isDora"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::isDora"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::tpSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule6mInDimE", "tensorrt_llm::runtime::LoraModule::mInDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule11mInDimFirstE", "tensorrt_llm::runtime::LoraModule::mInDimFirst"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule13mInTpSplitDimE", "tensorrt_llm::runtime::LoraModule::mInTpSplitDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule7mOutDimE", "tensorrt_llm::runtime::LoraModule::mOutDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12mOutDimFirstE", "tensorrt_llm::runtime::LoraModule::mOutDimFirst"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule14mOutTpSplitDimE", "tensorrt_llm::runtime::LoraModule::mOutTpSplitDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule5mTypeE", "tensorrt_llm::runtime::LoraModule::mType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule4nameEv", "tensorrt_llm::runtime::LoraModule::name"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", "tensorrt_llm::runtime::LoraModule::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", "tensorrt_llm::runtime::LoraModule::operator=::o"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6outDimEv", "tensorrt_llm::runtime::LoraModule::outDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11outDimFirstEv", "tensorrt_llm::runtime::LoraModule::outDimFirst"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::outSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::outSize::adapterSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule13outTpSplitDimEv", "tensorrt_llm::runtime::LoraModule::outTpSplitDim"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", "tensorrt_llm::runtime::LoraModule::toModuleName"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", "tensorrt_llm::runtime::LoraModule::toModuleName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", "tensorrt_llm::runtime::LoraModule::toModuleName::id"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", "tensorrt_llm::runtime::LoraModule::toModuleName::t"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", "tensorrt_llm::runtime::LoraModule::toModuleType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", "tensorrt_llm::runtime::LoraModule::toModuleType::name"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5valueEv", "tensorrt_llm::runtime::LoraModule::value"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14LoraTaskIdTypeE", "tensorrt_llm::runtime::LoraTaskIdType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", "tensorrt_llm::runtime::MPI_group_barrier"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", "tensorrt_llm::runtime::MPI_group_barrier::ranks"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModuleE", "tensorrt_llm::runtime::MedusaModule"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule13MedusaChoicesE", "tensorrt_llm::runtime::MedusaModule::MedusaChoices"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleEv", "tensorrt_llm::runtime::MedusaModule::MedusaModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule::maxAcceptedTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule::maxDraftTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule9TensorPtrE", "tensorrt_llm::runtime::MedusaModule::TensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12MedusaModule16getMedusaChoicesEv", "tensorrt_llm::runtime::MedusaModule::getMedusaChoices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule21mDefaultMedusaChoicesE", "tensorrt_llm::runtime::MedusaModule::mDefaultMedusaChoices"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCountersE", "tensorrt_llm::runtime::MemoryCounters"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8DiffTypeE", "tensorrt_llm::runtime::MemoryCounters::DiffType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters14MemoryCountersEv", "tensorrt_llm::runtime::MemoryCounters::MemoryCounters"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10SizeType32E", "tensorrt_llm::runtime::MemoryCounters::SizeType32"], [1, 3, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::T"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::memoryType"], [1, 4, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString::bytes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString::bytes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString::precision"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString::precision"], [1, 3, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::T"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::memoryType"], [1, 4, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::size"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getCpuEv", "tensorrt_llm::runtime::MemoryCounters::getCpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getCpuDiffEv", "tensorrt_llm::runtime::MemoryCounters::getCpuDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getGpuEv", "tensorrt_llm::runtime::MemoryCounters::getGpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getGpuDiffEv", "tensorrt_llm::runtime::MemoryCounters::getGpuDiff"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11getInstanceEv", "tensorrt_llm::runtime::MemoryCounters::getInstance"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters9getPinnedEv", "tensorrt_llm::runtime::MemoryCounters::getPinned"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedDiffEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedPoolEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedPool"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters17getPinnedPoolDiffEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedPoolDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getUVMEv", "tensorrt_llm::runtime::MemoryCounters::getUVM"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getUVMDiffEv", "tensorrt_llm::runtime::MemoryCounters::getUVMDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mCpuE", "tensorrt_llm::runtime::MemoryCounters::mCpu"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mCpuDiffE", "tensorrt_llm::runtime::MemoryCounters::mCpuDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mGpuE", "tensorrt_llm::runtime::MemoryCounters::mGpu"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mGpuDiffE", "tensorrt_llm::runtime::MemoryCounters::mGpuDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters7mPinnedE", "tensorrt_llm::runtime::MemoryCounters::mPinned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedDiffE", "tensorrt_llm::runtime::MemoryCounters::mPinnedDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedPoolE", "tensorrt_llm::runtime::MemoryCounters::mPinnedPool"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters15mPinnedPoolDiffE", "tensorrt_llm::runtime::MemoryCounters::mPinnedPoolDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mUVME", "tensorrt_llm::runtime::MemoryCounters::mUVM"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mUVMDiffE", "tensorrt_llm::runtime::MemoryCounters::mUVMDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters8toStringEv", "tensorrt_llm::runtime::MemoryCounters::toString"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryTypeE", "tensorrt_llm::runtime::MemoryType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kCPUE", "tensorrt_llm::runtime::MemoryType::kCPU"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kGPUE", "tensorrt_llm::runtime::MemoryType::kGPU"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType7kPINNEDE", "tensorrt_llm::runtime::MemoryType::kPINNED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType11kPINNEDPOOLE", "tensorrt_llm::runtime::MemoryType::kPINNEDPOOL"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kUVME", "tensorrt_llm::runtime::MemoryType::kUVM"], [1, 2, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", "tensorrt_llm::runtime::MemoryTypeString"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", "tensorrt_llm::runtime::MemoryTypeString::T"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kCPU>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kCPU>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kGPU>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kGPU>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNED>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNED>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNEDPOOL>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNEDPOOL>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kUVM>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kUVM>::value"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfigE", "tensorrt_llm::runtime::ModelConfig"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheTypeE", "tensorrt_llm::runtime::ModelConfig::KVCacheType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType11kCONTINUOUSE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kCONTINUOUS"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType9kDISABLEDE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kDISABLED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType6kPAGEDE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kPAGED"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", "tensorrt_llm::runtime::ModelConfig::KVCacheTypeFromString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", "tensorrt_llm::runtime::ModelConfig::KVCacheTypeFromString::value"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerTypeE", "tensorrt_llm::runtime::ModelConfig::LayerType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kATTENTIONE", "tensorrt_llm::runtime::ModelConfig::LayerType::kATTENTION"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType7kLINEARE", "tensorrt_llm::runtime::ModelConfig::LayerType::kLINEAR"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType5kNOOPE", "tensorrt_llm::runtime::ModelConfig::LayerType::kNOOP"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kRECURRENTE", "tensorrt_llm::runtime::ModelConfig::LayerType::kRECURRENT"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsTypeE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType9kDisabledE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType::kDisabled"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType8kEnabledE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType::kEnabled"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbAttentionLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbRnnLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::vocabSize"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariantE", "tensorrt_llm::runtime::ModelConfig::ModelVariant"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant8kChatGlmE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kChatGlm"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant7kEncDecE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kEncDec"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGlmE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kGlm"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGptE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kGpt"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant6kMambaE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kMamba"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant15kRecurrentGemmaE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kRecurrentGemma"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfigE", "tensorrt_llm::runtime::ModelConfig::RnnConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig10convKernelE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::convKernel"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig14rnnConvDimSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnConvDimSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig11rnnHeadSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnHeadSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig13rnnHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig9stateSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::stateSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeContextLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEv", "tensorrt_llm::runtime::ModelConfig::computeContextLogits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeContextLogits::computeContextLogits"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEv", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits::computeGenerationLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::layerType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::layerType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig32disableSeamlessLookaheadDecodingEv", "tensorrt_llm::runtime::ModelConfig::disableSeamlessLookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", "tensorrt_llm::runtime::ModelConfig::enableSeamlessLookaheadDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", "tensorrt_llm::runtime::ModelConfig::enableSeamlessLookaheadDecoding::maxDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getContextFMHAEv", "tensorrt_llm::runtime::ModelConfig::getContextFMHA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getDataTypeEv", "tensorrt_llm::runtime::ModelConfig::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getEncoderHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getEncoderHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getGemmAllReduceDtypeEv", "tensorrt_llm::runtime::ModelConfig::getGemmAllReduceDtype"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getKVCacheTypeEv", "tensorrt_llm::runtime::ModelConfig::getKVCacheType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getKvDataTypeEv", "tensorrt_llm::runtime::ModelConfig::getKvDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getLayerTypesEv", "tensorrt_llm::runtime::ModelConfig::getLayerTypes"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLogitsDtypeEv", "tensorrt_llm::runtime::ModelConfig::getLogitsDtype"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLoraModulesEv", "tensorrt_llm::runtime::ModelConfig::getLoraModules"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getManageWeightsTypeEv", "tensorrt_llm::runtime::ModelConfig::getManageWeightsType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBatchSizeEv", "tensorrt_llm::runtime::ModelConfig::getMaxBatchSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBeamWidthEv", "tensorrt_llm::runtime::ModelConfig::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig25getMaxDecodingDraftTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxDecodingDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getMaxDecodingTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxDecodingTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMaxEncoderLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxEncoderLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxInputLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxInputLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxLoraRankEv", "tensorrt_llm::runtime::ModelConfig::getMaxLoraRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxNumTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxNumTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24getMaxPositionEmbeddingsEv", "tensorrt_llm::runtime::ModelConfig::getMaxPositionEmbeddings"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig30getMaxPromptEmbeddingTableSizeEv", "tensorrt_llm::runtime::ModelConfig::getMaxPromptEmbeddingTableSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getMaxSequenceLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxSequenceLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMlpHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getMlpHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getModelNameEv", "tensorrt_llm::runtime::ModelConfig::getModelName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getModelVariantEv", "tensorrt_llm::runtime::ModelConfig::getModelVariant"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10getNbHeadsEv", "tensorrt_llm::runtime::ModelConfig::getNbHeads"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbKvHeads"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbKvHeads::layerIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbLayers::pipelineParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getNumKvHeadsPerLayerEv", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::isCrossAttention"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getNumLanguagesEv", "tensorrt_llm::runtime::ModelConfig::getNumLanguages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig25getOptProfilesSplitPointsEv", "tensorrt_llm::runtime::ModelConfig::getOptProfilesSplitPoints"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19getPagedContextFMHAEv", "tensorrt_llm::runtime::ModelConfig::getPagedContextFMHA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getPpReduceScatterEv", "tensorrt_llm::runtime::ModelConfig::getPpReduceScatter"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getQuantModeEv", "tensorrt_llm::runtime::ModelConfig::getQuantMode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getRnnConfigEv", "tensorrt_llm::runtime::ModelConfig::getRnnConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getRotaryEmbeddingDimEv", "tensorrt_llm::runtime::ModelConfig::getRotaryEmbeddingDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getSizePerHeadEv", "tensorrt_llm::runtime::ModelConfig::getSizePerHead"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig26getSpeculativeDecodingModeEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingMode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28getSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModulePtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModulePtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::isCrossAttention"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getTokensPerBlockEv", "tensorrt_llm::runtime::ModelConfig::getTokensPerBlock"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getVocabSizeEv", "tensorrt_llm::runtime::ModelConfig::getVocabSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getVocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getVocabSizePadded::worldSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12hasRnnConfigEv", "tensorrt_llm::runtime::ModelConfig::hasRnnConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28hasSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::hasSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19isContinuousKVCacheEv", "tensorrt_llm::runtime::ModelConfig::isContinuousKVCache"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16isKVCacheEnabledEv", "tensorrt_llm::runtime::ModelConfig::isKVCacheEnabled"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12isMultiModalEv", "tensorrt_llm::runtime::ModelConfig::isMultiModal"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14isPagedKVCacheEv", "tensorrt_llm::runtime::ModelConfig::isPagedKVCache"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10isRnnBasedEv", "tensorrt_llm::runtime::ModelConfig::isRnnBased"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18isTransformerBasedEv", "tensorrt_llm::runtime::ModelConfig::isTransformerBased"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig9isWhisperEv", "tensorrt_llm::runtime::ModelConfig::isWhisper"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig29kDEFAULT_NUM_TOKENS_PER_BLOCKE", "tensorrt_llm::runtime::ModelConfig::kDEFAULT_NUM_TOKENS_PER_BLOCK"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26kOPT_PROFILES_SPLIT_POINTSE", "tensorrt_llm::runtime::ModelConfig::kOPT_PROFILES_SPLIT_POINTS"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mComputeContextLogitsE", "tensorrt_llm::runtime::ModelConfig::mComputeContextLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mComputeGenerationLogitsE", "tensorrt_llm::runtime::ModelConfig::mComputeGenerationLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mContextFMHAE", "tensorrt_llm::runtime::ModelConfig::mContextFMHA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mDataTypeE", "tensorrt_llm::runtime::ModelConfig::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mEncoderHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mEncoderHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mGemmAllReduceDtypeE", "tensorrt_llm::runtime::ModelConfig::mGemmAllReduceDtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mInputPackedE", "tensorrt_llm::runtime::ModelConfig::mInputPacked"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mKVCacheTypeE", "tensorrt_llm::runtime::ModelConfig::mKVCacheType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mLayerTypesE", "tensorrt_llm::runtime::ModelConfig::mLayerTypes"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLogitsDtypeE", "tensorrt_llm::runtime::ModelConfig::mLogitsDtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLoraModulesE", "tensorrt_llm::runtime::ModelConfig::mLoraModules"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mManageWeightsTypeE", "tensorrt_llm::runtime::ModelConfig::mManageWeightsType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBatchSizeE", "tensorrt_llm::runtime::ModelConfig::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBeamWidthE", "tensorrt_llm::runtime::ModelConfig::mMaxBeamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMaxEncoderLenE", "tensorrt_llm::runtime::ModelConfig::mMaxEncoderLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxInputLenE", "tensorrt_llm::runtime::ModelConfig::mMaxInputLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxLoraRankE", "tensorrt_llm::runtime::ModelConfig::mMaxLoraRank"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxNumTokensE", "tensorrt_llm::runtime::ModelConfig::mMaxNumTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mMaxPositionEmbeddingsE", "tensorrt_llm::runtime::ModelConfig::mMaxPositionEmbeddings"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mMaxPromptEmbeddingTableSizeE", "tensorrt_llm::runtime::ModelConfig::mMaxPromptEmbeddingTableSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mMaxSequenceLenE", "tensorrt_llm::runtime::ModelConfig::mMaxSequenceLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMlpHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mMlpHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mModelNameE", "tensorrt_llm::runtime::ModelConfig::mModelName"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mModelVariantE", "tensorrt_llm::runtime::ModelConfig::mModelVariant"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mNbAttentionLayersE", "tensorrt_llm::runtime::ModelConfig::mNbAttentionLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig8mNbHeadsE", "tensorrt_llm::runtime::ModelConfig::mNbHeads"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mNbLayersE", "tensorrt_llm::runtime::ModelConfig::mNbLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mNbRnnLayersE", "tensorrt_llm::runtime::ModelConfig::mNbRnnLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mNumKvHeadsPerAttentionLayerE", "tensorrt_llm::runtime::ModelConfig::mNumKvHeadsPerAttentionLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig33mNumKvHeadsPerCrossAttentionLayerE", "tensorrt_llm::runtime::ModelConfig::mNumKvHeadsPerCrossAttentionLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mNumLanguagesE", "tensorrt_llm::runtime::ModelConfig::mNumLanguages"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17mPagedContextFMHAE", "tensorrt_llm::runtime::ModelConfig::mPagedContextFMHA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mPagedStateE", "tensorrt_llm::runtime::ModelConfig::mPagedState"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16mPpReduceScatterE", "tensorrt_llm::runtime::ModelConfig::mPpReduceScatter"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mQuantModeE", "tensorrt_llm::runtime::ModelConfig::mQuantMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mRnnConfigE", "tensorrt_llm::runtime::ModelConfig::mRnnConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mRotaryEmbeddingDimE", "tensorrt_llm::runtime::ModelConfig::mRotaryEmbeddingDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mSizePerHeadE", "tensorrt_llm::runtime::ModelConfig::mSizePerHead"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20mSkipCrossAttnBlocksE", "tensorrt_llm::runtime::ModelConfig::mSkipCrossAttnBlocks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mSpeculativeDecodingModeE", "tensorrt_llm::runtime::ModelConfig::mSpeculativeDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26mSpeculativeDecodingModuleE", "tensorrt_llm::runtime::ModelConfig::mSpeculativeDecodingModule"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mTokensPerBlockE", "tensorrt_llm::runtime::ModelConfig::mTokensPerBlock"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseCrossAttentionE", "tensorrt_llm::runtime::ModelConfig::mUseCrossAttention"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23mUseGemmAllReducePluginE", "tensorrt_llm::runtime::ModelConfig::mUseGemmAllReducePlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseGptAttentionPluginE", "tensorrt_llm::runtime::ModelConfig::mUseGptAttentionPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mUseLoraPluginE", "tensorrt_llm::runtime::ModelConfig::mUseLoraPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUseMambaConv1dPluginE", "tensorrt_llm::runtime::ModelConfig::mUseMambaConv1dPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mUseMropeE", "tensorrt_llm::runtime::ModelConfig::mUseMrope"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUsePositionEmbeddingE", "tensorrt_llm::runtime::ModelConfig::mUsePositionEmbedding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseShapeInferenceE", "tensorrt_llm::runtime::ModelConfig::mUseShapeInference"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseTokenTypeEmbeddingE", "tensorrt_llm::runtime::ModelConfig::mUseTokenTypeEmbedding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mVocabSizeE", "tensorrt_llm::runtime::ModelConfig::mVocabSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30resetSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::resetSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setContextFMHA"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setContextFMHA::contextFMHA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setEncoderHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setEncoderHiddenSize::encoderHiddenSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setGemmAllReduceDtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setGemmAllReduceDtype::inputDtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", "tensorrt_llm::runtime::ModelConfig::setKVCacheType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", "tensorrt_llm::runtime::ModelConfig::setKVCacheType::kvCacheType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", "tensorrt_llm::runtime::ModelConfig::setLayerTypes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", "tensorrt_llm::runtime::ModelConfig::setLayerTypes::layerTypes"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setLogitsDtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setLogitsDtype::inputDtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", "tensorrt_llm::runtime::ModelConfig::setLoraModules"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", "tensorrt_llm::runtime::ModelConfig::setLoraModules::loraModules"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", "tensorrt_llm::runtime::ModelConfig::setManageWeightsType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", "tensorrt_llm::runtime::ModelConfig::setManageWeightsType::manageWeightType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBatchSize::maxBatchSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBeamWidth::maxBeamWidth"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxEncoderLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxEncoderLen::maxEncoderLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxInputLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxInputLen::maxInputLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxLoraRank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxLoraRank::maxLoraRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setMaxNumTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setMaxNumTokens::maxNumTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPositionEmbeddings"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPositionEmbeddings::maxPositionEmbeddings"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPromptEmbeddingTableSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPromptEmbeddingTableSize::maxPromptEmbeddingTableSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxSequenceLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxSequenceLen::maxSequenceLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMlpHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMlpHiddenSize::mlpHiddenSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", "tensorrt_llm::runtime::ModelConfig::setModelName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", "tensorrt_llm::runtime::ModelConfig::setModelName::modelName"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", "tensorrt_llm::runtime::ModelConfig::setModelVariant"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", "tensorrt_llm::runtime::ModelConfig::setModelVariant::modelVariant"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbCrossKvHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbCrossKvHeads::nbKvHeads"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbKvHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbKvHeads::nbKvHeads"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerCrossLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerCrossLayer::headsPerLayer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerLayer::headsPerLayer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumLanguages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumLanguages::numLanguages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setPagedContextFMHA"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setPagedContextFMHA::pagedContextFMHA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", "tensorrt_llm::runtime::ModelConfig::setPpReduceScatter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", "tensorrt_llm::runtime::ModelConfig::setPpReduceScatter::ppReduceScatter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", "tensorrt_llm::runtime::ModelConfig::setQuantMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", "tensorrt_llm::runtime::ModelConfig::setQuantMode::QuantMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", "tensorrt_llm::runtime::ModelConfig::setRnnConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", "tensorrt_llm::runtime::ModelConfig::setRnnConfig::rnnConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setRotaryEmbeddingDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setRotaryEmbeddingDim::rotaryEmbeddingDim"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setSizePerHead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setSizePerHead::sizePerHead"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", "tensorrt_llm::runtime::ModelConfig::setSkipCrossAttnBlocks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", "tensorrt_llm::runtime::ModelConfig::setSkipCrossAttnBlocks::skipCrossAttnBlocks"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingMode::mode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingModule::speculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setTokensPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setTokensPerBlock::TokensPerBlock"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", "tensorrt_llm::runtime::ModelConfig::setUseCrossAttention"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", "tensorrt_llm::runtime::ModelConfig::setUseCrossAttention::useCrossAttention"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", "tensorrt_llm::runtime::ModelConfig::setUseMrope"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", "tensorrt_llm::runtime::ModelConfig::setUseMrope::useMrope"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUsePositionEmbedding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUsePositionEmbedding::usePositionEmbedding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", "tensorrt_llm::runtime::ModelConfig::setUseShapeInference"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", "tensorrt_llm::runtime::ModelConfig::setUseShapeInference::useShapeInference"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUseTokenTypeEmbedding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUseTokenTypeEmbedding::useTokenTypeEmbedding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19skipCrossAttnBlocksEv", "tensorrt_llm::runtime::ModelConfig::skipCrossAttnBlocks"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24supportsInflightBatchingEv", "tensorrt_llm::runtime::ModelConfig::supportsInflightBatching"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useCrossAttentionEv", "tensorrt_llm::runtime::ModelConfig::useCrossAttention"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEv", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin::useGemmAllReducePlugin"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEv", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin::useGptAttentionPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18useLanguageAdapterEv", "tensorrt_llm::runtime::ModelConfig::useLanguageAdapter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13useLoraPluginEv", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin::useLoraPlugin"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEv", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin::useMambaConv1dPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig8useMropeEv", "tensorrt_llm::runtime::ModelConfig::useMrope"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", "tensorrt_llm::runtime::ModelConfig::usePackedInput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14usePackedInputEv", "tensorrt_llm::runtime::ModelConfig::usePackedInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", "tensorrt_llm::runtime::ModelConfig::usePackedInput::inputPacked"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", "tensorrt_llm::runtime::ModelConfig::usePagedState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13usePagedStateEv", "tensorrt_llm::runtime::ModelConfig::usePagedState"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", "tensorrt_llm::runtime::ModelConfig::usePagedState::pagedState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20usePositionEmbeddingEv", "tensorrt_llm::runtime::ModelConfig::usePositionEmbedding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15usePromptTuningEv", "tensorrt_llm::runtime::ModelConfig::usePromptTuning"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useShapeInferenceEv", "tensorrt_llm::runtime::ModelConfig::useShapeInference"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useTokenTypeEmbeddingEv", "tensorrt_llm::runtime::ModelConfig::useTokenTypeEmbedding"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", "tensorrt_llm::runtime::PointerElementType"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", "tensorrt_llm::runtime::PointerElementType::T"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParamsE", "tensorrt_llm::runtime::PromptTuningParams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::embeddingTable"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::tasks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::vocabSize"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams10SizeType32E", "tensorrt_llm::runtime::PromptTuningParams::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams9TensorPtrE", "tensorrt_llm::runtime::PromptTuningParams::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::numContextRequests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::packedInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::reqBeamWidths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::reqPromptLengths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::tasksHost"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngineE", "tensorrt_llm::runtime::RawEngine"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", "tensorrt_llm::runtime::RawEngine::AddressWithSize"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", "tensorrt_llm::runtime::RawEngine::FilePath"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", "tensorrt_llm::runtime::RawEngine::HostMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineAddr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineBuffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::RawEngine::enginePath"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineSize"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4TypeE", "tensorrt_llm::runtime::RawEngine::Type"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", "tensorrt_llm::runtime::RawEngine::Type::AddressWithSize"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", "tensorrt_llm::runtime::RawEngine::Type::FilePath"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", "tensorrt_llm::runtime::RawEngine::Type::HostMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getAddressEv", "tensorrt_llm::runtime::RawEngine::getAddress"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine13getHostMemoryEv", "tensorrt_llm::runtime::RawEngine::getHostMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine23getManagedWeightsMapOptEv", "tensorrt_llm::runtime::RawEngine::getManagedWeightsMapOpt"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getPathEv", "tensorrt_llm::runtime::RawEngine::getPath"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getPathOptEv", "tensorrt_llm::runtime::RawEngine::getPathOpt"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getSizeEv", "tensorrt_llm::runtime::RawEngine::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getTypeEv", "tensorrt_llm::runtime::RawEngine::getType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineAddrE", "tensorrt_llm::runtime::RawEngine::mEngineAddr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine13mEngineBufferE", "tensorrt_llm::runtime::RawEngine::mEngineBuffer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEnginePathE", "tensorrt_llm::runtime::RawEngine::mEnginePath"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineSizeE", "tensorrt_llm::runtime::RawEngine::mEngineSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine18mManagedWeightsMapE", "tensorrt_llm::runtime::RawEngine::mManagedWeightsMap"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine5mTypeE", "tensorrt_llm::runtime::RawEngine::mType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", "tensorrt_llm::runtime::RawEngine::setManagedWeightsMap"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", "tensorrt_llm::runtime::RawEngine::setManagedWeightsMap::managedWeightsMap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::setPath"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::setPath::enginePath"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11RequestTypeE", "tensorrt_llm::runtime::RequestType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11RequestType8kCONTEXTE", "tensorrt_llm::runtime::RequestType::kCONTEXT"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11RequestType11kGENERATIONE", "tensorrt_llm::runtime::RequestType::kGENERATION"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::runtime::RuntimeDefaults"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsEv", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults::maxAttentionWindowVec"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults::sinkTokenLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults21maxAttentionWindowVecE", "tensorrt_llm::runtime::RuntimeDefaults::maxAttentionWindowVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15sinkTokenLengthE", "tensorrt_llm::runtime::RuntimeDefaults::sinkTokenLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfigE", "tensorrt_llm::runtime::SamplingConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9FloatTypeE", "tensorrt_llm::runtime::SamplingConfig::FloatType"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", "tensorrt_llm::runtime::SamplingConfig::OptVec"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", "tensorrt_llm::runtime::SamplingConfig::OptVec::T"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::beamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::configs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::externalDraftTokensConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::samplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig23beamSearchDiversityRateE", "tensorrt_llm::runtime::SamplingConfig::beamSearchDiversityRate"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9beamWidthE", "tensorrt_llm::runtime::SamplingConfig::beamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14beamWidthArrayE", "tensorrt_llm::runtime::SamplingConfig::beamWidthArray"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11cumLogProbsE", "tensorrt_llm::runtime::SamplingConfig::cumLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig24draftAcceptanceThresholdE", "tensorrt_llm::runtime::SamplingConfig::draftAcceptanceThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13earlyStoppingE", "tensorrt_llm::runtime::SamplingConfig::earlyStopping"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig16frequencyPenaltyE", "tensorrt_llm::runtime::SamplingConfig::frequencyPenalty"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::accessor"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::configs"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::defaultValue"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig15getMaxBeamWidthEv", "tensorrt_llm::runtime::SamplingConfig::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig17getNumReturnBeamsEv", "tensorrt_llm::runtime::SamplingConfig::getNumReturnBeams"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13lengthPenaltyE", "tensorrt_llm::runtime::SamplingConfig::lengthPenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9minLengthE", "tensorrt_llm::runtime::SamplingConfig::minLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4minPE", "tensorrt_llm::runtime::SamplingConfig::minP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17noRepeatNgramSizeE", "tensorrt_llm::runtime::SamplingConfig::noRepeatNgramSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17normalizeLogProbsE", "tensorrt_llm::runtime::SamplingConfig::normalizeLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig18numReturnSequencesE", "tensorrt_llm::runtime::SamplingConfig::numReturnSequences"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::runtime::SamplingConfig::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::runtime::SamplingConfig::operator==::other"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig19originalTemperatureE", "tensorrt_llm::runtime::SamplingConfig::originalTemperature"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14outputLogProbsE", "tensorrt_llm::runtime::SamplingConfig::outputLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15presencePenaltyE", "tensorrt_llm::runtime::SamplingConfig::presencePenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig10randomSeedE", "tensorrt_llm::runtime::SamplingConfig::randomSeed"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17repetitionPenaltyE", "tensorrt_llm::runtime::SamplingConfig::repetitionPenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11temperatureE", "tensorrt_llm::runtime::SamplingConfig::temperature"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topKE", "tensorrt_llm::runtime::SamplingConfig::topK"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15topKMedusaHeadsE", "tensorrt_llm::runtime::SamplingConfig::topKMedusaHeads"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topPE", "tensorrt_llm::runtime::SamplingConfig::topP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9topPDecayE", "tensorrt_llm::runtime::SamplingConfig::topPDecay"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig7topPMinE", "tensorrt_llm::runtime::SamplingConfig::topPMin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig12topPResetIdsE", "tensorrt_llm::runtime::SamplingConfig::topPResetIds"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::defaultValue"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::vec"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig8validateEv", "tensorrt_llm::runtime::SamplingConfig::validate"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::max"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::min"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::name"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::vec"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10SizeType32E", "tensorrt_llm::runtime::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10SizeType64E", "tensorrt_llm::runtime::SizeType64"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingModeE", "tensorrt_llm::runtime::SpeculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19DraftTokensExternalEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::DraftTokensExternal"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5EagleEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::Eagle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19ExplicitDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::ExplicitDraftTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode17LookaheadDecodingEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::LookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6MedusaEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::Medusa"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode4NoneEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::None"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::SpeculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::SpeculativeDecodingMode::state"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode14UnderlyingTypeE", "tensorrt_llm::runtime::SpeculativeDecodingMode::UnderlyingType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::allBitSet"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::allBitSet::bits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::anyBitSet"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::anyBitSet::bits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode14hasDraftLogitsEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::hasDraftLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isDraftTokensExternalEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isDraftTokensExternal"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode7isEagleEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isEagle"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isExplicitDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isExplicitDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19isLookaheadDecodingEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isLookaheadDecoding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode8isMedusaEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isMedusa"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode6isNoneEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isNone"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kDraftTokensExternalE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kDraftTokensExternal"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6kEagleE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kEagle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kExplicitDraftTokensE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kExplicitDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode18kLookaheadDecodingE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kLookaheadDecoding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode7kMedusaE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kMedusa"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5kNoneE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kNone"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6mStateE", "tensorrt_llm::runtime::SpeculativeDecodingMode::mState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode20needsDecoderPrologueEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::needsDecoderPrologue"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18needsKVCacheRewindEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::needsKVCacheRewind"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", "tensorrt_llm::runtime::SpeculativeDecodingMode::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", "tensorrt_llm::runtime::SpeculativeDecodingMode::operator==::other"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19predictsDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::predictsDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21requiresAttentionMaskEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::requiresAttentionMask"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18updatesPositionIdsEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::updatesPositionIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19variableDraftLengthEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::variableDraftLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleE", "tensorrt_llm::runtime::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxNumPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::o"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule21computeNumPackedMasksEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::computeNumPackedMasks"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule25getMaxDecodingDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDecodingDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule20getMaxDecodingTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDecodingTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule18getMaxDraftPathLenEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDraftPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule14getMaxNumPathsEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxNumPaths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule13getMaxPathLenEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule17getNumPackedMasksEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getNumPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule23mMaxDecodingDraftTokensE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxDecodingDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule16mMaxDraftPathLenE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxDraftPathLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18mMaxNumPackedMasksE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxNumPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule12mMaxNumPathsE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxNumPaths"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::operator=::o"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftPathLen::maxDraftPathLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftTokens::maxDraftTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxNumPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxNumPaths::maxNumPaths"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleD0Ev", "tensorrt_llm::runtime::SpeculativeDecodingModule::~SpeculativeDecodingModule"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", "tensorrt_llm::runtime::StringPtrMap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", "tensorrt_llm::runtime::StringPtrMap::T"], [1, 2, 1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", "tensorrt_llm::runtime::TRTDataType"], [1, 8, 1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", "tensorrt_llm::runtime::TRTDataType::T"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", "tensorrt_llm::runtime::TRTDataType<T*>"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", "tensorrt_llm::runtime::TRTDataType<T*>::T"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE15kUnderlyingTypeE", "tensorrt_llm::runtime::TRTDataType<T*>::kUnderlyingType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE5valueE", "tensorrt_llm::runtime::TRTDataType<T*>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIbEE", "tensorrt_llm::runtime::TRTDataType<bool>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIbE5valueE", "tensorrt_llm::runtime::TRTDataType<bool>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIfEE", "tensorrt_llm::runtime::TRTDataType<float>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIfE5valueE", "tensorrt_llm::runtime::TRTDataType<float>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeI4halfEE", "tensorrt_llm::runtime::TRTDataType<half>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeI4halfE5valueE", "tensorrt_llm::runtime::TRTDataType<half>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEEE", "tensorrt_llm::runtime::TRTDataType<kernels::FinishedState>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEE5valueE", "tensorrt_llm::runtime::TRTDataType<kernels::FinishedState>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEEE", "tensorrt_llm::runtime::TRTDataType<kernels::KVCacheIndex>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEE5valueE", "tensorrt_llm::runtime::TRTDataType<kernels::KVCacheIndex>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEEE", "tensorrt_llm::runtime::TRTDataType<runtime::RequestType>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEE5valueE", "tensorrt_llm::runtime::TRTDataType<runtime::RequestType>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int32_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int32_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int64_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int64_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int8_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int8_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint32_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint32_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint64_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint64_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint8_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint8_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIPvEE", "tensorrt_llm::runtime::TRTDataType<void*>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIPvE5valueE", "tensorrt_llm::runtime::TRTDataType<void*>::value"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLoggerE", "tensorrt_llm::runtime::TllmLogger"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8getLevelEv", "tensorrt_llm::runtime::TllmLogger::getLevel"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log::msg"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log::severity"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", "tensorrt_llm::runtime::TllmLogger::setLevel"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", "tensorrt_llm::runtime::TllmLogger::setLevel::level"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16TokenExtraIdTypeE", "tensorrt_llm::runtime::TokenExtraIdType"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11TokenIdTypeE", "tensorrt_llm::runtime::TokenIdType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueTokenE", "tensorrt_llm::runtime::UniqueToken"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", "tensorrt_llm::runtime::UniqueToken::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", "tensorrt_llm::runtime::UniqueToken::operator==::other"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken12tokenExtraIdE", "tensorrt_llm::runtime::UniqueToken::tokenExtraId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken7tokenIdE", "tensorrt_llm::runtime::UniqueToken::tokenId"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16VecTokenExtraIdsE", "tensorrt_llm::runtime::VecTokenExtraIds"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime15VecUniqueTokensE", "tensorrt_llm::runtime::VecUniqueTokens"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfigE", "tensorrt_llm::runtime::WorldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::deviceIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::enableAttentionDP"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::rank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::tensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17enableAttentionDPEv", "tensorrt_llm::runtime::WorldConfig::enableAttentionDP"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getContextParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getContextParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getContextParallelismEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig9getDeviceEv", "tensorrt_llm::runtime::WorldConfig::getDevice"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getDeviceOf"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getDeviceOf::rank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig15getGpusPerGroupEv", "tensorrt_llm::runtime::WorldConfig::getGpusPerGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14getGpusPerNodeEv", "tensorrt_llm::runtime::WorldConfig::getGpusPerNode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getLastRankEv", "tensorrt_llm::runtime::WorldConfig::getLastRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig12getLocalRankEv", "tensorrt_llm::runtime::WorldConfig::getLocalRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getNodeRankEv", "tensorrt_llm::runtime::WorldConfig::getNodeRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getNodeRankOf"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getNodeRankOf::rank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig24getPipelineParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getPipelineParallelismEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getRankEv", "tensorrt_llm::runtime::WorldConfig::getRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getSizeEv", "tensorrt_llm::runtime::WorldConfig::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getTensorParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getTensorParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig20getTensorParallelismEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17isContextParallelEv", "tensorrt_llm::runtime::WorldConfig::isContextParallel"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isFirstContextParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstContextParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig27isFirstPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig25isFirstTensorParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstTensorParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isLastPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isLastPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig18isPipelineParallelEv", "tensorrt_llm::runtime::WorldConfig::isPipelineParallel"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig16isTensorParallelEv", "tensorrt_llm::runtime::WorldConfig::isTensorParallel"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19kDefaultGpusPerNodeE", "tensorrt_llm::runtime::WorldConfig::kDefaultGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19mContextParallelismE", "tensorrt_llm::runtime::WorldConfig::mContextParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig10mDeviceIdsE", "tensorrt_llm::runtime::WorldConfig::mDeviceIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mEnableAttentionDPE", "tensorrt_llm::runtime::WorldConfig::mEnableAttentionDP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig12mGpusPerNodeE", "tensorrt_llm::runtime::WorldConfig::mGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig20mPipelineParallelismE", "tensorrt_llm::runtime::WorldConfig::mPipelineParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig5mRankE", "tensorrt_llm::runtime::WorldConfig::mRank"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mTensorParallelismE", "tensorrt_llm::runtime::WorldConfig::mTensorParallelism"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::deviceIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::enableAttentionDP"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::tensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14validMpiConfigEv", "tensorrt_llm::runtime::WorldConfig::validMpiConfig"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast::buffer"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast::buffer"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::bufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::bufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalBufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalBufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalTensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalTensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::tensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::tensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", "tensorrt_llm::runtime::canAccessPeer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", "tensorrt_llm::runtime::canAccessPeer::worldConfig"], [1, 3, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast"], [1, 8, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::D"], [1, 8, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast::T"], [1, 4, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::ptr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast::ptr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoderE", "tensorrt_llm::runtime::decoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoderE", "tensorrt_llm::runtime::decoder"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffersE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::BeamSearchBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::BeamSearchBuffers::bufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers15mCumLogProbsTmpE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mCumLogProbsTmp"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7mNumSMsE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mNumSMs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers21mOutputBeamHypothesesE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mOutputBeamHypotheses"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape::maxSequenceLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderStateE", "tensorrt_llm::runtime::decoder::DecoderState"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState::dtype"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16DecodingInputPtrE", "tensorrt_llm::runtime::decoder::DecoderState::DecodingInputPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState17DecodingOutputPtrE", "tensorrt_llm::runtime::decoder::DecoderState::DecodingOutputPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13LlmRequestPtrE", "tensorrt_llm::runtime::decoder::DecoderState::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13RequestVectorE", "tensorrt_llm::runtime::decoder::DecoderState::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState9TensorPtrE", "tensorrt_llm::runtime::decoder::DecoderState::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::speculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", "tensorrt_llm::runtime::decoder::DecoderState::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", "tensorrt_llm::runtime::decoder::DecoderState::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24getAcceptedLengthsCumSumEv", "tensorrt_llm::runtime::decoder::DecoderState::getAcceptedLengthsCumSum"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getAcceptedPackedPathsEv", "tensorrt_llm::runtime::decoder::DecoderState::getAcceptedPackedPaths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getAllNewTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getAllNewTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getBeamSearchBuffersEv", "tensorrt_llm::runtime::decoder::DecoderState::getBeamSearchBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsEv", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishReasonsEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishReasons"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishedStepsEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishedSteps"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getFinishedSumEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishedSum"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getIds::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState21getJointDecodingInputEv", "tensorrt_llm::runtime::decoder::DecoderState::getJointDecodingInput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getJointDecodingOutputEv", "tensorrt_llm::runtime::decoder::DecoderState::getJointDecodingOutput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsEv", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBeamWidthEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState27getMaxDecodingDecoderTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxDecodingDecoderTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getMaxDecodingEngineTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxDecodingEngineTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getMaxSequenceLengthEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxSequenceLength"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getNextDraftTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getNextDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getNextDraftTokensLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getNextDraftTokensLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState12getParentIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getParentIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getPrevDraftTokensLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getPrevDraftTokensLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getSequenceLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getSpeculativeDecodingModeEv", "tensorrt_llm::runtime::decoder::DecoderState::getSpeculativeDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mBeamSearchBuffersE", "tensorrt_llm::runtime::decoder::DecoderState::mBeamSearchBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState14mFinishedStepsE", "tensorrt_llm::runtime::decoder::DecoderState::mFinishedSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState19mJointDecodingInputE", "tensorrt_llm::runtime::decoder::DecoderState::mJointDecodingInput"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState20mJointDecodingOutputE", "tensorrt_llm::runtime::decoder::DecoderState::mJointDecodingOutput"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBatchSizeE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBeamWidthE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxBeamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState25mMaxDecodingDecoderTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxDecodingDecoderTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mMaxDecodingEngineTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxDecodingEngineTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mMaxSequenceLengthE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxSequenceLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mNumDecodingEngineTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mNumDecodingEngineTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mSpeculativeDecodingModeE", "tensorrt_llm::runtime::decoder::DecoderState::mSpeculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens::batchIdx"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens::numTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupEagle"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupEagle::eagleBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupExplicitDraftTokens"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupExplicitDraftTokens::explicitDraftTokensBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", "tensorrt_llm::runtime::decoder::DecoderState::setupLookahead"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", "tensorrt_llm::runtime::decoder::DecoderState::setupLookahead::lookaheadDecodingBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::maxTokensPerEngineStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::speculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", "tensorrt_llm::runtime::decoder_batch"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", "tensorrt_llm::runtime::decoder_batch"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5InputE", "tensorrt_llm::runtime::decoder_batch::Input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", "tensorrt_llm::runtime::decoder_batch::Input::Input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", "tensorrt_llm::runtime::decoder_batch::Input::Input::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input::maxDecoderSteps"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input14TensorConstPtrE", "tensorrt_llm::runtime::decoder_batch::Input::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Input::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input10batchSlotsE", "tensorrt_llm::runtime::decoder_batch::Input::batchSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input22batchSlotsRequestOrderE", "tensorrt_llm::runtime::decoder_batch::Input::batchSlotsRequestOrder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input16cacheIndirectionE", "tensorrt_llm::runtime::decoder_batch::Input::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input11eagleInputsE", "tensorrt_llm::runtime::decoder_batch::Input::eagleInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15eagleLastInputsE", "tensorrt_llm::runtime::decoder_batch::Input::eagleLastInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input25explicitDraftTokensInputsE", "tensorrt_llm::runtime::decoder_batch::Input::explicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input29explicitDraftTokensLastInputsE", "tensorrt_llm::runtime::decoder_batch::Input::explicitDraftTokensLastInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15generationStepsE", "tensorrt_llm::runtime::decoder_batch::Input::generationSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input6logitsE", "tensorrt_llm::runtime::decoder_batch::Input::logits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15maxDecoderStepsE", "tensorrt_llm::runtime::decoder_batch::Input::maxDecoderSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input20predictedDraftLogitsE", "tensorrt_llm::runtime::decoder_batch::Input::predictedDraftLogits"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6OutputE", "tensorrt_llm::runtime::decoder_batch::Output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output6OutputEv", "tensorrt_llm::runtime::decoder_batch::Output::Output"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Output::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output16cacheIndirectionE", "tensorrt_llm::runtime::decoder_batch::Output::cacheIndirection"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7RequestE", "tensorrt_llm::runtime::decoder_batch::Request"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9BufferPtrE", "tensorrt_llm::runtime::decoder_batch::Request::BufferPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::endId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::ids"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::inputLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::maxNewTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request14TensorConstPtrE", "tensorrt_llm::runtime::decoder_batch::Request::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Request::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12badWordsListE", "tensorrt_llm::runtime::decoder_batch::Request::badWordsList"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftLogitsE", "tensorrt_llm::runtime::decoder_batch::Request::draftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftTokensE", "tensorrt_llm::runtime::decoder_batch::Request::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5dtypeE", "tensorrt_llm::runtime::decoder_batch::Request::dtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11eagleConfigE", "tensorrt_llm::runtime::decoder_batch::Request::eagleConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13embeddingBiasE", "tensorrt_llm::runtime::decoder_batch::Request::embeddingBias"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5endIdE", "tensorrt_llm::runtime::decoder_batch::Request::endId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request28generatedTokensPerEngineStepE", "tensorrt_llm::runtime::decoder_batch::Request::generatedTokensPerEngineStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request3idsE", "tensorrt_llm::runtime::decoder_batch::Request::ids"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request8inputLenE", "tensorrt_llm::runtime::decoder_batch::Request::inputLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request22lookaheadRuntimeConfigE", "tensorrt_llm::runtime::decoder_batch::Request::lookaheadRuntimeConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12maxNewTokensE", "tensorrt_llm::runtime::decoder_batch::Request::maxNewTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11medusaPathsE", "tensorrt_llm::runtime::decoder_batch::Request::medusaPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13medusaTreeIdsE", "tensorrt_llm::runtime::decoder_batch::Request::medusaTreeIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13stopWordsListE", "tensorrt_llm::runtime::decoder_batch::Request::stopWordsList"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", "tensorrt_llm::runtime::getDefaultBatchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", "tensorrt_llm::runtime::getDefaultBatchSlots::batchSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", "tensorrt_llm::runtime::ipcNvlsFree"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", "tensorrt_llm::runtime::ipcNvlsFree::handle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime16ipcNvlsSupportedEv", "tensorrt_llm::runtime::ipcNvlsSupported"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_0"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_1"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_2"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<::c"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<::module"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<::os"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<::os"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::to_string"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::to_string"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::to_string::c"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::to_string::v"], [82, 9, 0, "-", "tensorrt_llm"]], "tensorrt_llm": [[77, 9, 0, "-", "functional"], [79, 9, 0, "-", "models"], [80, 9, 0, "-", "plugin"], [81, 9, 0, "-", "quantization"], [82, 9, 0, "-", "runtime"]], "tensorrt_llm.functional": [[77, 10, 1, "", "AllReduceFusionOp"], [77, 10, 1, "", "AllReduceParams"], [77, 10, 1, "", "AllReduceStrategy"], [77, 10, 1, "", "AttentionMaskType"], [77, 10, 1, "", "Conditional"], [77, 10, 1, "", "DimRange"], [77, 10, 1, "", "LayerNormPositionType"], [77, 10, 1, "", "LayerNormType"], [77, 10, 1, "", "MLPType"], [77, 10, 1, "", "PositionEmbeddingType"], [77, 10, 1, "", "RopeEmbeddingUtils"], [77, 10, 1, "", "RotaryScalingType"], [77, 10, 1, "", "SideStreamIDType"], [77, 10, 1, "", "SliceInputType"], [77, 10, 1, "", "Tensor"], [77, 14, 1, "", "abs"], [77, 14, 1, "", "activation"], [77, 14, 1, "", "add"], [77, 14, 1, "", "allgather"], [77, 14, 1, "", "allreduce"], [77, 14, 1, "", "arange"], [77, 14, 1, "", "argmax"], [77, 14, 1, "", "assertion"], [77, 14, 1, "", "avg_pool2d"], [77, 14, 1, "", "bert_attention"], [77, 14, 1, "", "broadcast_helper"], [77, 14, 1, "", "cast"], [77, 14, 1, "", "categorical_sample"], [77, 14, 1, "", "chunk"], [77, 14, 1, "", "clip"], [77, 14, 1, "", "concat"], [77, 14, 1, "", "constant"], [77, 14, 1, "", "constant_to_tensor_"], [77, 14, 1, "", "constants_to_tensors_"], [77, 14, 1, "", "conv1d"], [77, 14, 1, "", "conv2d"], [77, 14, 1, "", "conv3d"], [77, 14, 1, "", "conv_transpose2d"], [77, 14, 1, "", "cos"], [77, 14, 1, "", "cp_split_plugin"], [77, 14, 1, "", "create_allreduce_plugin"], [77, 14, 1, "", "cuda_stream_sync"], [77, 14, 1, "", "cumsum"], [77, 14, 1, "", "div"], [77, 14, 1, "", "dora_plugin"], [77, 14, 1, "", "einsum"], [77, 14, 1, "", "elementwise_binary"], [77, 14, 1, "", "embedding"], [77, 14, 1, "", "eq"], [77, 14, 1, "", "exp"], [77, 14, 1, "", "expand"], [77, 14, 1, "", "expand_dims"], [77, 14, 1, "", "expand_dims_like"], [77, 14, 1, "", "expand_mask"], [77, 14, 1, "", "flatten"], [77, 14, 1, "", "flip"], [77, 14, 1, "", "floordiv"], [77, 14, 1, "", "gather"], [77, 14, 1, "", "gather_last_token_logits"], [77, 14, 1, "", "gather_nd"], [77, 14, 1, "", "gegelu"], [77, 14, 1, "", "geglu"], [77, 14, 1, "", "gelu"], [77, 14, 1, "", "gemm_allreduce"], [77, 14, 1, "", "gemm_swiglu"], [77, 14, 1, "", "generate_alibi_biases"], [77, 14, 1, "", "generate_alibi_slopes"], [77, 14, 1, "", "generate_logn_scaling"], [77, 14, 1, "", "gpt_attention"], [77, 14, 1, "", "group_norm"], [77, 14, 1, "", "gt"], [77, 14, 1, "", "identity"], [77, 14, 1, "", "index_select"], [77, 14, 1, "", "int_clip"], [77, 14, 1, "", "interpolate"], [77, 14, 1, "", "is_gated_activation"], [77, 14, 1, "", "layer_norm"], [77, 14, 1, "", "log"], [77, 14, 1, "", "log_softmax"], [77, 14, 1, "", "lora_plugin"], [77, 14, 1, "", "low_latency_gemm"], [77, 14, 1, "", "low_latency_gemm_swiglu"], [77, 14, 1, "", "lt"], [77, 14, 1, "", "mamba_conv1d"], [77, 14, 1, "", "masked_scatter"], [77, 14, 1, "", "masked_select"], [77, 14, 1, "", "matmul"], [77, 14, 1, "", "max"], [77, 14, 1, "", "maximum"], [77, 14, 1, "", "mean"], [77, 14, 1, "", "meshgrid2d"], [77, 14, 1, "", "min"], [77, 14, 1, "", "minimum"], [77, 14, 1, "", "modulo"], [77, 14, 1, "", "mul"], [77, 14, 1, "", "non_gated_version"], [77, 14, 1, "", "nonzero"], [77, 14, 1, "", "not_op"], [77, 14, 1, "", "op_and"], [77, 14, 1, "", "op_or"], [77, 14, 1, "", "op_xor"], [77, 14, 1, "", "outer"], [77, 14, 1, "", "pad"], [77, 14, 1, "", "permute"], [77, 14, 1, "", "pow"], [77, 14, 1, "", "prod"], [77, 14, 1, "", "quick_gelu"], [77, 14, 1, "", "rand"], [77, 14, 1, "", "rearrange"], [77, 14, 1, "", "recv"], [77, 14, 1, "", "reduce"], [77, 14, 1, "", "reduce_scatter"], [77, 14, 1, "", "relu"], [77, 14, 1, "", "repeat"], [77, 14, 1, "", "repeat_interleave"], [77, 14, 1, "", "rg_lru"], [77, 14, 1, "", "rms_norm"], [77, 14, 1, "", "round"], [77, 14, 1, "", "scatter"], [77, 14, 1, "", "scatter_nd"], [77, 14, 1, "", "select"], [77, 14, 1, "", "selective_scan"], [77, 14, 1, "", "send"], [77, 14, 1, "", "shape"], [77, 14, 1, "", "sigmoid"], [77, 14, 1, "", "silu"], [77, 14, 1, "", "sin"], [77, 14, 1, "", "slice"], [77, 14, 1, "", "softmax"], [77, 14, 1, "", "softplus"], [77, 14, 1, "", "split"], [77, 14, 1, "", "sqrt"], [77, 14, 1, "", "squared_relu"], [77, 14, 1, "", "squeeze"], [77, 14, 1, "", "stack"], [77, 14, 1, "", "sub"], [77, 14, 1, "", "sum"], [77, 14, 1, "", "swiglu"], [77, 14, 1, "", "tanh"], [77, 14, 1, "", "topk"], [77, 14, 1, "", "transpose"], [77, 14, 1, "", "unary"], [77, 14, 1, "", "unbind"], [77, 14, 1, "", "unsqueeze"], [77, 14, 1, "", "view"], [77, 14, 1, "", "where"]], "tensorrt_llm.functional.AllReduceFusionOp": [[77, 11, 1, "", "LAST_PROCESS_FOR_UB"], [77, 11, 1, "", "MOE_ALLREDUCE_RESIDUAL_RMS_NORM"], [77, 11, 1, "", "NONE"], [77, 11, 1, "", "RESIDUAL_RMS_NORM"], [77, 11, 1, "", "RESIDUAL_RMS_NORM_OUT_QUANT_FP8"], [77, 11, 1, "", "RESIDUAL_RMS_NORM_OUT_QUANT_NVFP4"], [77, 11, 1, "", "RESIDUAL_RMS_NORM_QUANT_FP8"], [77, 11, 1, "", "RESIDUAL_RMS_NORM_QUANT_NVFP4"], [77, 11, 1, "", "RESIDUAL_RMS_PREPOST_NORM"]], "tensorrt_llm.functional.AllReduceParams": [[77, 12, 1, "", "has_affine"], [77, 12, 1, "", "has_bias"], [77, 12, 1, "", "has_scale"], [77, 12, 1, "", "update_strategy"]], "tensorrt_llm.functional.AllReduceStrategy": [[77, 11, 1, "", "AUTO"], [77, 11, 1, "", "MIN_LATENCY"], [77, 11, 1, "", "NCCL"], [77, 11, 1, "", "ONESHOT"], [77, 11, 1, "", "TWOSHOT"], [77, 11, 1, "", "UB"]], "tensorrt_llm.functional.AttentionMaskType": [[77, 11, 1, "", "bidirectional"], [77, 11, 1, "", "bidirectionalglm"], [77, 11, 1, "", "blocksparse"], [77, 11, 1, "", "causal"], [77, 11, 1, "", "custom_mask"], [77, 11, 1, "", "padding"], [77, 11, 1, "", "sliding_window_causal"]], "tensorrt_llm.functional.Conditional": [[77, 12, 1, "", "add_input"], [77, 12, 1, "", "add_output"]], "tensorrt_llm.functional.LayerNormPositionType": [[77, 11, 1, "", "post_layernorm"], [77, 11, 1, "", "pre_layernorm"]], "tensorrt_llm.functional.LayerNormType": [[77, 11, 1, "", "GroupNorm"], [77, 11, 1, "", "LayerNorm"], [77, 11, 1, "", "RmsNorm"]], "tensorrt_llm.functional.MLPType": [[77, 11, 1, "", "FusedGatedMLP"], [77, 11, 1, "", "GatedMLP"], [77, 11, 1, "", "MLP"]], "tensorrt_llm.functional.PositionEmbeddingType": [[77, 11, 1, "", "alibi"], [77, 11, 1, "", "alibi_with_scale"], [77, 11, 1, "", "chatglm"], [77, 12, 1, "", "choices"], [77, 11, 1, "", "deferred"], [77, 12, 1, "", "from_string"], [77, 12, 1, "", "is_alibi"], [77, 12, 1, "", "is_deferred"], [77, 12, 1, "", "is_mrope"], [77, 12, 1, "", "is_rope"], [77, 11, 1, "", "learned_absolute"], [77, 11, 1, "", "long_rope"], [77, 11, 1, "", "mrope"], [77, 11, 1, "", "relative"], [77, 11, 1, "", "rope_gpt_neox"], [77, 11, 1, "", "rope_gptj"], [77, 11, 1, "", "yarn"]], "tensorrt_llm.functional.RopeEmbeddingUtils": [[77, 12, 1, "", "apply_llama3_scaling"], [77, 12, 1, "", "apply_rotary_pos_emb"], [77, 12, 1, "", "apply_rotary_pos_emb_chatglm"], [77, 12, 1, "", "apply_rotary_pos_emb_cogvlm"], [77, 12, 1, "", "create_fake_weight"], [77, 12, 1, "", "create_sinusoidal_positions"], [77, 12, 1, "", "create_sinusoidal_positions_for_attention_plugin"], [77, 12, 1, "", "create_sinusoidal_positions_for_cogvlm_attention_plugin"], [77, 12, 1, "", "create_sinusoidal_positions_long_rope"], [77, 12, 1, "", "create_sinusoidal_positions_yarn"], [77, 12, 1, "", "rotate_every_two"], [77, 12, 1, "", "rotate_half"]], "tensorrt_llm.functional.RotaryScalingType": [[77, 11, 1, "", "dynamic"], [77, 12, 1, "", "from_string"], [77, 11, 1, "", "linear"], [77, 11, 1, "", "llama3"], [77, 11, 1, "", "longrope"], [77, 11, 1, "", "mrope"], [77, 11, 1, "", "none"], [77, 11, 1, "", "yarn"]], "tensorrt_llm.functional.SideStreamIDType": [[77, 11, 1, "", "disable"], [77, 11, 1, "", "moe"]], "tensorrt_llm.functional.SliceInputType": [[77, 11, 1, "", "axes"], [77, 11, 1, "", "data"], [77, 11, 1, "", "fill_value"], [77, 11, 1, "", "size"], [77, 11, 1, "", "start"], [77, 11, 1, "", "stride"]], "tensorrt_llm.functional.Tensor": [[77, 12, 1, "", "abs"], [77, 12, 1, "", "cast"], [77, 13, 1, "", "dtype"], [77, 12, 1, "", "flatten"], [77, 12, 1, "", "get_parent"], [77, 12, 1, "", "get_users"], [77, 12, 1, "", "is_dynamic"], [77, 12, 1, "", "is_trt_wrapper"], [77, 13, 1, "", "location"], [77, 12, 1, "", "log"], [77, 12, 1, "", "mark_output"], [77, 12, 1, "", "max"], [77, 12, 1, "", "mean"], [77, 13, 1, "", "name"], [77, 12, 1, "", "ndim"], [77, 13, 1, "", "network"], [77, 12, 1, "", "permute"], [77, 12, 1, "", "rank"], [77, 12, 1, "", "repeat"], [77, 12, 1, "", "replace_all_uses_with"], [77, 12, 1, "", "select"], [77, 13, 1, "", "shape"], [77, 12, 1, "", "size"], [77, 12, 1, "", "split"], [77, 12, 1, "", "sqrt"], [77, 12, 1, "", "squeeze"], [77, 12, 1, "", "transpose"], [77, 12, 1, "", "unbind"], [77, 12, 1, "", "unsqueeze"], [77, 12, 1, "", "view"]], "tensorrt_llm.layers": [[78, 9, 0, "-", "activation"], [78, 9, 0, "-", "attention"], [78, 9, 0, "-", "cast"], [78, 9, 0, "-", "conv"], [78, 9, 0, "-", "embedding"], [78, 9, 0, "-", "linear"], [78, 9, 0, "-", "mlp"], [78, 9, 0, "-", "normalization"], [78, 9, 0, "-", "pooling"]], "tensorrt_llm.layers.activation": [[78, 10, 1, "", "Mish"]], "tensorrt_llm.layers.activation.Mish": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention": [[78, 10, 1, "", "Attention"], [78, 10, 1, "", "AttentionMaskParams"], [78, 10, 1, "", "AttentionParams"], [78, 10, 1, "", "BertAttention"], [78, 10, 1, "", "BlockSparseAttnParams"], [78, 10, 1, "", "CogVLMAttention"], [78, 10, 1, "", "DeepseekV2Attention"], [78, 10, 1, "", "DiffusersAttention"], [78, 10, 1, "", "KeyValueCacheParams"], [78, 10, 1, "", "MropeParams"], [78, 10, 1, "", "SpecDecodingParams"], [78, 14, 1, "", "compute_relative_bias"], [78, 14, 1, "", "make_causal_mask"]], "tensorrt_llm.layers.attention.Attention": [[78, 12, 1, "", "create_attention_const_params"], [78, 12, 1, "", "fill_attention_params"], [78, 12, 1, "", "forward"], [78, 12, 1, "", "postprocess"], [78, 12, 1, "", "set_rel_attn_table"]], "tensorrt_llm.layers.attention.AttentionParams": [[78, 12, 1, "", "fill_attention_const_params_for_long_rope"], [78, 12, 1, "", "fill_attention_const_params_for_rope"], [78, 12, 1, "", "is_valid"], [78, 12, 1, "", "is_valid_cross_attn"]], "tensorrt_llm.layers.attention.BertAttention": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention.CogVLMAttention": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention.DeepseekV2Attention": [[78, 12, 1, "", "forward"], [78, 12, 1, "", "postprocess"], [78, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.attention.DiffusersAttention": [[78, 12, 1, "", "forward"], [78, 12, 1, "", "joint_attn_forward"]], "tensorrt_llm.layers.attention.KeyValueCacheParams": [[78, 12, 1, "", "fill_none_tensor_list"], [78, 12, 1, "", "get_first_past_key_value"], [78, 12, 1, "", "is_valid"]], "tensorrt_llm.layers.cast": [[78, 10, 1, "", "Cast"]], "tensorrt_llm.layers.cast.Cast": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv": [[78, 10, 1, "", "Conv1d"], [78, 10, 1, "", "Conv2d"], [78, 10, 1, "", "Conv3d"], [78, 10, 1, "", "ConvTranspose2d"]], "tensorrt_llm.layers.conv.Conv1d": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.Conv2d": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.Conv3d": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.ConvTranspose2d": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding": [[78, 10, 1, "", "CombinedTimestepLabelEmbeddings"], [78, 10, 1, "", "CombinedTimestepTextProjEmbeddings"], [78, 10, 1, "", "Embedding"], [78, 10, 1, "", "LabelEmbedding"], [78, 10, 1, "", "PixArtAlphaTextProjection"], [78, 10, 1, "", "PromptTuningEmbedding"], [78, 10, 1, "", "SD3PatchEmbed"], [78, 10, 1, "", "TimestepEmbedding"], [78, 10, 1, "", "Timesteps"], [78, 14, 1, "", "get_1d_sincos_pos_embed_from_grid"], [78, 14, 1, "", "get_2d_sincos_pos_embed"], [78, 14, 1, "", "get_2d_sincos_pos_embed_from_grid"], [78, 14, 1, "", "get_timestep_embedding"]], "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.Embedding": [[78, 12, 1, "", "forward"], [78, 12, 1, "", "postprocess"], [78, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.embedding.LabelEmbedding": [[78, 12, 1, "", "forward"], [78, 12, 1, "", "token_drop"]], "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.PromptTuningEmbedding": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.SD3PatchEmbed": [[78, 12, 1, "", "cropped_pos_embed"], [78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.TimestepEmbedding": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.Timesteps": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.linear": [[78, 11, 1, "", "ColumnLinear"], [78, 10, 1, "", "Linear"], [78, 10, 1, "", "LinearBase"], [78, 10, 1, "", "RowLinear"]], "tensorrt_llm.layers.linear.Linear": [[78, 12, 1, "", "collect_and_bias"], [78, 12, 1, "", "postprocess"], [78, 12, 1, "", "tp_split_dim"]], "tensorrt_llm.layers.linear.LinearBase": [[78, 12, 1, "", "collect_and_bias"], [78, 12, 1, "", "forward"], [78, 12, 1, "", "get_weight"], [78, 12, 1, "", "multiply_and_lora"], [78, 12, 1, "", "multiply_collect"], [78, 12, 1, "", "tp_split_dim"], [78, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.linear.RowLinear": [[78, 12, 1, "", "collect_and_bias"], [78, 12, 1, "", "multiply_collect"], [78, 12, 1, "", "tp_split_dim"]], "tensorrt_llm.layers.mlp": [[78, 10, 1, "", "FusedGatedMLP"], [78, 10, 1, "", "GatedMLP"], [78, 10, 1, "", "LinearActivation"], [78, 10, 1, "", "LinearApproximateGELU"], [78, 10, 1, "", "LinearGEGLU"], [78, 10, 1, "", "LinearGELU"], [78, 10, 1, "", "LinearSwiGLU"], [78, 10, 1, "", "MLP"], [78, 14, 1, "", "fc_gate_dora"], [78, 14, 1, "", "fc_gate_lora"]], "tensorrt_llm.layers.mlp.FusedGatedMLP": [[78, 12, 1, "", "fc_gate"], [78, 12, 1, "", "fc_gate_plugin"], [78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.GatedMLP": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearActivation": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearApproximateGELU": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearGEGLU": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearGELU": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearSwiGLU": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.MLP": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization": [[78, 10, 1, "", "AdaLayerNorm"], [78, 10, 1, "", "AdaLayerNormContinuous"], [78, 10, 1, "", "AdaLayerNormZero"], [78, 10, 1, "", "AdaLayerNormZeroSingle"], [78, 10, 1, "", "GroupNorm"], [78, 10, 1, "", "LayerNorm"], [78, 10, 1, "", "RmsNorm"], [78, 10, 1, "", "SD35AdaLayerNormZeroX"]], "tensorrt_llm.layers.normalization.AdaLayerNorm": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormContinuous": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormZero": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.GroupNorm": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.LayerNorm": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.RmsNorm": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX": [[78, 12, 1, "", "forward"]], "tensorrt_llm.layers.pooling": [[78, 10, 1, "", "AvgPool2d"]], "tensorrt_llm.layers.pooling.AvgPool2d": [[78, 12, 1, "", "forward"]], "tensorrt_llm.llmapi": [[65, 10, 1, "", "BatchingType"], [65, 10, 1, "", "BuildCacheConfig"], [65, 10, 1, "", "BuildConfig"], [65, 10, 1, "", "CacheTransceiverConfig"], [65, 10, 1, "", "CalibConfig"], [65, 10, 1, "", "CapacitySchedulerPolicy"], [65, 10, 1, "", "CompletionOutput"], [65, 10, 1, "", "ContextChunkingPolicy"], [65, 10, 1, "", "DisaggregatedParams"], [65, 10, 1, "", "DynamicBatchConfig"], [65, 10, 1, "", "EagleDecodingConfig"], [65, 10, 1, "", "ExtendedRuntimePerfKnobConfig"], [65, 10, 1, "", "GuidedDecodingParams"], [65, 10, 1, "", "KvCacheConfig"], [65, 10, 1, "", "KvCacheRetentionConfig"], [65, 10, 1, "", "LLM"], [65, 10, 1, "", "LookaheadDecodingConfig"], [65, 10, 1, "", "MTPDecodingConfig"], [65, 10, 1, "", "MedusaDecodingConfig"], [65, 10, 1, "", "MpiCommSession"], [65, 10, 1, "", "QuantAlgo"], [65, 10, 1, "", "QuantConfig"], [65, 10, 1, "", "RequestError"], [65, 10, 1, "", "RequestOutput"], [65, 10, 1, "", "SamplingParams"], [65, 10, 1, "", "SchedulerConfig"]], "tensorrt_llm.llmapi.BatchingType": [[65, 11, 1, "", "INFLIGHT"], [65, 11, 1, "", "STATIC"]], "tensorrt_llm.llmapi.BuildCacheConfig": [[65, 12, 1, "", "__init__"], [65, 13, 1, "id7", "cache_root"], [65, 13, 1, "id8", "max_cache_storage_gb"], [65, 13, 1, "id9", "max_records"]], "tensorrt_llm.llmapi.BuildConfig": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "auto_parallel_config"], [65, 11, 1, "", "dry_run"], [65, 11, 1, "", "enable_debug_output"], [65, 11, 1, "", "force_num_profiles"], [65, 12, 1, "", "from_dict"], [65, 12, 1, "", "from_json_file"], [65, 11, 1, "", "gather_context_logits"], [65, 11, 1, "", "gather_generation_logits"], [65, 11, 1, "", "input_timing_cache"], [65, 11, 1, "", "kv_cache_type"], [65, 11, 1, "", "lora_config"], [65, 11, 1, "", "max_batch_size"], [65, 11, 1, "", "max_beam_width"], [65, 11, 1, "", "max_draft_len"], [65, 11, 1, "", "max_encoder_input_len"], [65, 11, 1, "", "max_input_len"], [65, 11, 1, "", "max_num_tokens"], [65, 11, 1, "", "max_prompt_embedding_table_size"], [65, 11, 1, "", "max_seq_len"], [65, 11, 1, "", "monitor_memory"], [65, 11, 1, "", "opt_batch_size"], [65, 11, 1, "", "opt_num_tokens"], [65, 11, 1, "", "output_timing_cache"], [65, 11, 1, "", "plugin_config"], [65, 11, 1, "", "profiling_verbosity"], [65, 11, 1, "", "speculative_decoding_mode"], [65, 11, 1, "", "strongly_typed"], [65, 12, 1, "", "to_dict"], [65, 12, 1, "", "update"], [65, 12, 1, "", "update_from_dict"], [65, 12, 1, "", "update_kv_cache_type"], [65, 11, 1, "", "use_mrope"], [65, 11, 1, "", "use_refit"], [65, 11, 1, "", "use_strip_plan"], [65, 11, 1, "", "visualize_network"], [65, 11, 1, "", "weight_sparsity"], [65, 11, 1, "", "weight_streaming"]], "tensorrt_llm.llmapi.CacheTransceiverConfig": [[65, 15, 1, "", "max_num_tokens"], [65, 11, 1, "", "model_config"]], "tensorrt_llm.llmapi.CalibConfig": [[65, 15, 1, "", "calib_batch_size"], [65, 15, 1, "", "calib_batches"], [65, 15, 1, "", "calib_dataset"], [65, 15, 1, "", "calib_max_seq_length"], [65, 15, 1, "", "device"], [65, 12, 1, "", "from_dict"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "random_seed"], [65, 12, 1, "", "to_dict"], [65, 15, 1, "", "tokenizer_max_seq_length"]], "tensorrt_llm.llmapi.CapacitySchedulerPolicy": [[65, 11, 1, "", "GUARANTEED_NO_EVICT"], [65, 11, 1, "", "MAX_UTILIZATION"], [65, 11, 1, "", "STATIC_BATCH"]], "tensorrt_llm.llmapi.CompletionOutput": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "cumulative_logprob"], [65, 11, 1, "", "disaggregated_params"], [65, 11, 1, "", "finish_reason"], [65, 11, 1, "", "generation_logits"], [65, 11, 1, "", "index"], [65, 13, 1, "id2", "length"], [65, 11, 1, "", "logprobs"], [65, 13, 1, "id3", "logprobs_diff"], [65, 11, 1, "", "prompt_logprobs"], [65, 11, 1, "", "stop_reason"], [65, 11, 1, "", "text"], [65, 13, 1, "id4", "text_diff"], [65, 11, 1, "", "token_ids"], [65, 13, 1, "id5", "token_ids_diff"]], "tensorrt_llm.llmapi.ContextChunkingPolicy": [[65, 11, 1, "", "EQUAL_PROGRESS"], [65, 11, 1, "", "FIRST_COME_FIRST_SERVED"]], "tensorrt_llm.llmapi.DisaggregatedParams": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "ctx_request_id"], [65, 11, 1, "", "draft_tokens"], [65, 11, 1, "", "first_gen_tokens"], [65, 12, 1, "", "get_context_phase_params"], [65, 12, 1, "", "get_request_type"], [65, 11, 1, "", "opaque_state"], [65, 11, 1, "", "request_type"]], "tensorrt_llm.llmapi.DynamicBatchConfig": [[65, 15, 1, "", "dynamic_batch_moving_average_window"], [65, 15, 1, "", "enable_batch_size_tuning"], [65, 15, 1, "", "enable_max_num_tokens_tuning"], [65, 11, 1, "", "model_config"]], "tensorrt_llm.llmapi.EagleDecodingConfig": [[65, 11, 1, "", "decoding_type"], [65, 15, 1, "", "dynamic_tree_max_topK"], [65, 15, 1, "", "eagle_choices"], [65, 12, 1, "", "from_dict"], [65, 15, 1, "", "greedy_sampling"], [65, 15, 1, "", "max_non_leaves_per_layer"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "num_eagle_layers"], [65, 15, 1, "", "posterior_threshold"], [65, 15, 1, "", "pytorch_eagle_weights_path"], [65, 15, 1, "", "use_dynamic_tree"]], "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig": [[65, 15, 1, "", "cuda_graph_cache_size"], [65, 15, 1, "", "cuda_graph_mode"], [65, 15, 1, "", "enable_context_fmha_fp32_acc"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "multi_block_mode"]], "tensorrt_llm.llmapi.GuidedDecodingParams": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "grammar"], [65, 11, 1, "", "json"], [65, 11, 1, "", "json_object"], [65, 11, 1, "", "regex"], [65, 11, 1, "", "structural_tag"]], "tensorrt_llm.llmapi.KvCacheConfig": [[65, 15, 1, "", "copy_on_partial_reuse"], [65, 15, 1, "", "cross_kv_cache_fraction"], [65, 15, 1, "", "enable_block_reuse"], [65, 15, 1, "", "enable_partial_reuse"], [65, 15, 1, "", "event_buffer_max_size"], [65, 15, 1, "", "free_gpu_memory_fraction"], [65, 15, 1, "", "host_cache_size"], [65, 15, 1, "", "max_attention_window"], [65, 15, 1, "", "max_tokens"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "onboard_blocks"], [65, 15, 1, "", "secondary_offload_min_priority"], [65, 15, 1, "", "sink_token_length"]], "tensorrt_llm.llmapi.KvCacheRetentionConfig": [[65, 10, 1, "", "TokenRangeRetentionConfig"], [65, 12, 1, "", "__init__"], [65, 13, 1, "", "decode_duration_ms"], [65, 13, 1, "", "decode_retention_priority"], [65, 13, 1, "", "token_range_retention_configs"]], "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig": [[65, 12, 1, "", "__init__"], [65, 13, 1, "", "duration_ms"], [65, 13, 1, "", "priority"], [65, 13, 1, "", "token_end"], [65, 13, 1, "", "token_start"]], "tensorrt_llm.llmapi.LLM": [[65, 12, 1, "", "__init__"], [65, 12, 1, "", "generate"], [65, 12, 1, "", "generate_async"], [65, 12, 1, "", "get_kv_cache_events"], [65, 12, 1, "", "get_kv_cache_events_async"], [65, 12, 1, "", "get_stats"], [65, 12, 1, "", "get_stats_async"], [65, 12, 1, "", "save"], [65, 12, 1, "", "shutdown"], [65, 13, 1, "id0", "tokenizer"], [65, 13, 1, "id1", "workspace"]], "tensorrt_llm.llmapi.LookaheadDecodingConfig": [[65, 12, 1, "", "__init__"], [65, 12, 1, "", "calculate_speculative_resource"], [65, 11, 1, "", "decoding_type"], [65, 12, 1, "", "from_dict"], [65, 15, 1, "", "max_ngram_size"], [65, 15, 1, "", "max_verification_set_size"], [65, 15, 1, "", "max_window_size"], [65, 11, 1, "", "model_config"], [65, 12, 1, "", "validate_positive_values"]], "tensorrt_llm.llmapi.MTPDecodingConfig": [[65, 11, 1, "", "decoding_type"], [65, 12, 1, "", "from_dict"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "num_nextn_predict_layers"], [65, 15, 1, "", "relaxed_delta"], [65, 15, 1, "", "relaxed_topk"], [65, 15, 1, "", "use_relaxed_acceptance_for_thinking"]], "tensorrt_llm.llmapi.MedusaDecodingConfig": [[65, 11, 1, "", "decoding_type"], [65, 12, 1, "", "from_dict"], [65, 15, 1, "", "medusa_choices"], [65, 11, 1, "", "model_config"], [65, 15, 1, "", "num_medusa_heads"]], "tensorrt_llm.llmapi.MpiCommSession": [[65, 12, 1, "", "__init__"], [65, 12, 1, "", "abort"], [65, 12, 1, "", "get_comm"], [65, 12, 1, "", "shutdown"], [65, 12, 1, "", "submit"], [65, 12, 1, "", "submit_sync"]], "tensorrt_llm.llmapi.QuantAlgo": [[65, 11, 1, "", "FP8"], [65, 11, 1, "", "FP8_BLOCK_SCALES"], [65, 11, 1, "", "FP8_PER_CHANNEL_PER_TOKEN"], [65, 11, 1, "", "INT8"], [65, 11, 1, "", "MIXED_PRECISION"], [65, 11, 1, "", "NO_QUANT"], [65, 11, 1, "", "NVFP4"], [65, 11, 1, "", "W4A16"], [65, 11, 1, "", "W4A16_AWQ"], [65, 11, 1, "", "W4A16_GPTQ"], [65, 11, 1, "", "W4A8_AWQ"], [65, 11, 1, "", "W4A8_QSERVE_PER_CHANNEL"], [65, 11, 1, "", "W4A8_QSERVE_PER_GROUP"], [65, 11, 1, "", "W8A16"], [65, 11, 1, "", "W8A16_GPTQ"], [65, 11, 1, "", "W8A8_SQ_PER_CHANNEL"], [65, 11, 1, "", "W8A8_SQ_PER_CHANNEL_PER_TENSOR_PLUGIN"], [65, 11, 1, "", "W8A8_SQ_PER_CHANNEL_PER_TOKEN_PLUGIN"], [65, 11, 1, "", "W8A8_SQ_PER_TENSOR_PER_TOKEN_PLUGIN"], [65, 11, 1, "", "W8A8_SQ_PER_TENSOR_PLUGIN"]], "tensorrt_llm.llmapi.QuantConfig": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "clamp_val"], [65, 11, 1, "", "exclude_modules"], [65, 12, 1, "", "from_dict"], [65, 11, 1, "", "group_size"], [65, 11, 1, "", "has_zero_point"], [65, 12, 1, "", "is_module_excluded_from_quantization"], [65, 11, 1, "", "kv_cache_quant_algo"], [65, 13, 1, "", "layer_quant_mode"], [65, 11, 1, "", "pre_quant_scale"], [65, 11, 1, "", "quant_algo"], [65, 13, 1, "", "quant_mode"], [65, 11, 1, "", "smoothquant_val"], [65, 12, 1, "", "to_dict"], [65, 11, 1, "", "use_meta_recipe"]], "tensorrt_llm.llmapi.RequestOutput": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "context_logits"], [65, 11, 1, "", "finished"], [65, 11, 1, "", "outputs"], [65, 13, 1, "id6", "prompt"], [65, 11, 1, "", "prompt_token_ids"], [65, 11, 1, "", "request_id"]], "tensorrt_llm.llmapi.SamplingParams": [[65, 12, 1, "", "__init__"], [65, 11, 1, "", "add_special_tokens"], [65, 11, 1, "", "additional_model_outputs"], [65, 11, 1, "", "apply_batched_logits_processor"], [65, 11, 1, "", "bad"], [65, 11, 1, "", "bad_token_ids"], [65, 11, 1, "", "beam_search_diversity_rate"], [65, 11, 1, "", "beam_width_array"], [65, 11, 1, "", "best_of"], [65, 11, 1, "", "detokenize"], [65, 11, 1, "", "early_stopping"], [65, 11, 1, "", "embedding_bias"], [65, 11, 1, "", "end_id"], [65, 11, 1, "", "exclude_input_from_output"], [65, 11, 1, "", "frequency_penalty"], [65, 11, 1, "", "guided_decoding"], [65, 11, 1, "", "ignore_eos"], [65, 11, 1, "", "include_stop_str_in_output"], [65, 11, 1, "", "length_penalty"], [65, 11, 1, "", "logits_processor"], [65, 11, 1, "", "logprobs"], [65, 11, 1, "", "lookahead_config"], [65, 11, 1, "", "max_tokens"], [65, 11, 1, "", "min_p"], [65, 11, 1, "", "min_tokens"], [65, 11, 1, "", "n"], [65, 11, 1, "", "no_repeat_ngram_size"], [65, 11, 1, "", "pad_id"], [65, 11, 1, "", "presence_penalty"], [65, 11, 1, "", "prompt_logprobs"], [65, 11, 1, "", "repetition_penalty"], [65, 11, 1, "", "return_context_logits"], [65, 11, 1, "", "return_encoder_output"], [65, 11, 1, "", "return_generation_logits"], [65, 11, 1, "", "return_perf_metrics"], [65, 11, 1, "", "seed"], [65, 11, 1, "", "skip_special_tokens"], [65, 11, 1, "", "spaces_between_special_tokens"], [65, 11, 1, "", "stop"], [65, 11, 1, "", "stop_token_ids"], [65, 11, 1, "", "temperature"], [65, 11, 1, "", "top_k"], [65, 11, 1, "", "top_p"], [65, 11, 1, "", "top_p_decay"], [65, 11, 1, "", "top_p_min"], [65, 11, 1, "", "top_p_reset_ids"], [65, 11, 1, "", "truncate_prompt_tokens"], [65, 11, 1, "", "use_beam_search"]], "tensorrt_llm.llmapi.SchedulerConfig": [[65, 15, 1, "", "capacity_scheduler_policy"], [65, 15, 1, "", "context_chunking_policy"], [65, 15, 1, "", "dynamic_batch_config"], [65, 11, 1, "", "model_config"]], "tensorrt_llm.models": [[79, 10, 1, "", "BaichuanForCausalLM"], [79, 10, 1, "", "BertForQuestionAnswering"], [79, 10, 1, "", "BertForSequenceClassification"], [79, 10, 1, "", "BertModel"], [79, 10, 1, "", "BloomForCausalLM"], [79, 10, 1, "", "BloomModel"], [79, 10, 1, "", "CLIPVisionTransformer"], [79, 10, 1, "", "ChatGLMConfig"], [79, 10, 1, "", "ChatGLMForCausalLM"], [79, 10, 1, "", "ChatGLMModel"], [79, 10, 1, "", "CogVLMConfig"], [79, 10, 1, "", "CogVLMForCausalLM"], [79, 10, 1, "", "CohereForCausalLM"], [79, 10, 1, "", "DbrxConfig"], [79, 10, 1, "", "DbrxForCausalLM"], [79, 10, 1, "", "DecoderModel"], [79, 10, 1, "", "DeepseekForCausalLM"], [79, 10, 1, "", "DeepseekV2ForCausalLM"], [79, 10, 1, "", "DiT"], [79, 10, 1, "", "EagleForCausalLM"], [79, 10, 1, "", "EncoderModel"], [79, 10, 1, "", "FalconConfig"], [79, 10, 1, "", "FalconForCausalLM"], [79, 10, 1, "", "FalconModel"], [79, 10, 1, "", "GPTConfig"], [79, 10, 1, "", "GPTForCausalLM"], [79, 10, 1, "", "GPTJConfig"], [79, 10, 1, "", "GPTJForCausalLM"], [79, 10, 1, "", "GPTJModel"], [79, 10, 1, "", "GPTModel"], [79, 10, 1, "", "GPTNeoXForCausalLM"], [79, 10, 1, "", "GPTNeoXModel"], [79, 10, 1, "", "GemmaConfig"], [79, 10, 1, "", "GemmaForCausalLM"], [79, 10, 1, "", "LLaMAConfig"], [79, 10, 1, "", "LLaMAForCausalLM"], [79, 10, 1, "", "LLaMAModel"], [79, 10, 1, "", "LlavaNextVisionConfig"], [79, 10, 1, "", "LlavaNextVisionWrapper"], [79, 10, 1, "", "MLLaMAForCausalLM"], [79, 10, 1, "", "MPTForCausalLM"], [79, 10, 1, "", "MPTModel"], [79, 10, 1, "", "MambaForCausalLM"], [79, 10, 1, "", "MedusaConfig"], [79, 10, 1, "", "MedusaForCausalLm"], [79, 10, 1, "", "OPTForCausalLM"], [79, 10, 1, "", "OPTModel"], [79, 10, 1, "", "Phi3ForCausalLM"], [79, 10, 1, "", "Phi3Model"], [79, 10, 1, "", "PhiForCausalLM"], [79, 10, 1, "", "PhiModel"], [79, 10, 1, "", "PretrainedConfig"], [79, 10, 1, "", "PretrainedModel"], [79, 10, 1, "", "ReDrafterForCausalLM"], [79, 10, 1, "", "RecurrentGemmaForCausalLM"], [79, 11, 1, "", "RobertaForQuestionAnswering"], [79, 11, 1, "", "RobertaForSequenceClassification"], [79, 11, 1, "", "RobertaModel"], [79, 10, 1, "", "SD3Transformer2DModel"], [79, 10, 1, "", "SpeculativeDecodingMode"], [79, 10, 1, "", "WhisperEncoder"]], "tensorrt_llm.models.BaichuanForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "quantize"]], "tensorrt_llm.models.BertForQuestionAnswering": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.BertForSequenceClassification": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.BertModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.BloomModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.CLIPVisionTransformer": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.ChatGLMConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.ChatGLMForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "quantize"]], "tensorrt_llm.models.ChatGLMModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.CogVLMConfig": [[79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.CogVLMForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "default_plugin_config"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "quantize"]], "tensorrt_llm.models.CohereForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DbrxConfig": [[79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.DbrxForCausalLM": [[79, 11, 1, "", "config_class"]], "tensorrt_llm.models.DecoderModel": [[79, 12, 1, "", "check_config"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "precompute_relative_attention_bias"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.DeepseekForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DeepseekV2ForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DiT": [[79, 12, 1, "", "check_config"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "forward_with_cfg"], [79, 12, 1, "", "forward_without_cfg"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "unpatchify"]], "tensorrt_llm.models.EagleForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.EncoderModel": [[79, 12, 1, "", "check_config"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "precompute_relative_attention_bias"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "use_lora"], [79, 12, 1, "", "use_prompt_tuning"]], "tensorrt_llm.models.FalconConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.FalconForCausalLM": [[79, 12, 1, "", "check_config"], [79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.FalconModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "from_nemo"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GPTForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "from_nemo"], [79, 12, 1, "", "quantize"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.GPTJConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GPTJForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.GPTJModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTNeoXModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.GemmaConfig": [[79, 11, 1, "", "GEMMA2_ADDED_FIELDS"], [79, 11, 1, "", "GEMMA3_ADDED_FIELDS"], [79, 11, 1, "", "GEMMA_ADDED_FIELDS"], [79, 11, 1, "", "VERBATIM"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "gemma2_config"], [79, 12, 1, "", "gemma3_config"], [79, 12, 1, "", "get_hf_config"], [79, 13, 1, "", "is_gemma_2"], [79, 13, 1, "", "is_gemma_3"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GemmaForCausalLM": [[79, 11, 1, "", "NATIVE_QUANT_FLOW"], [79, 12, 1, "", "assert_valid_quant_algo"], [79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "quantize"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.LLaMAConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "from_meta_ckpt"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.LLaMAForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "default_plugin_config"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "from_meta_ckpt"], [79, 12, 1, "", "quantize"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.LLaMAModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.LlavaNextVisionConfig": [[79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.LlavaNextVisionWrapper": [[79, 12, 1, "", "forward"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "save_checkpoint"]], "tensorrt_llm.models.MLLaMAForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.MPTForCausalLM": [[79, 12, 1, "", "check_config"]], "tensorrt_llm.models.MPTModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.MambaForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.MedusaConfig": [[79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "to_dict"]], "tensorrt_llm.models.MedusaForCausalLm": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.OPTForCausalLM": [[79, 12, 1, "", "check_config"]], "tensorrt_llm.models.OPTModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.Phi3ForCausalLM": [[79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.Phi3Model": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.PhiForCausalLM": [[79, 12, 1, "", "check_config"], [79, 11, 1, "", "config_class"], [79, 12, 1, "", "from_hugging_face"], [79, 12, 1, "", "use_lora"]], "tensorrt_llm.models.PhiModel": [[79, 12, 1, "", "forward"]], "tensorrt_llm.models.PretrainedConfig": [[79, 12, 1, "", "create_runtime_defaults"], [79, 12, 1, "", "for_each_rank"], [79, 12, 1, "", "from_checkpoint"], [79, 12, 1, "", "from_dict"], [79, 12, 1, "", "from_json_file"], [79, 12, 1, "", "get_config_group"], [79, 12, 1, "", "has_config_group"], [79, 13, 1, "", "kv_dtype"], [79, 13, 1, "", "quant_algo"], [79, 13, 1, "", "quant_mode"], [79, 12, 1, "", "set_if_not_exist"], [79, 12, 1, "", "set_rank"], [79, 12, 1, "", "to_dict"], [79, 12, 1, "", "to_json_file"], [79, 12, 1, "", "to_layer_quant_config"]], "tensorrt_llm.models.PretrainedModel": [[79, 12, 1, "", "check_config"], [79, 12, 1, "", "from_checkpoint"], [79, 12, 1, "", "from_config"], [79, 12, 1, "", "load"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "quantize"], [79, 12, 1, "", "release"], [79, 12, 1, "", "save_checkpoint"]], "tensorrt_llm.models.ReDrafterForCausalLM": [[79, 12, 1, "", "forward"], [79, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.RecurrentGemmaForCausalLM": [[79, 12, 1, "", "forward"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "prepare_recurrent_inputs"]], "tensorrt_llm.models.SD3Transformer2DModel": [[79, 13, 1, "", "attn_processors"], [79, 11, 1, "", "config_class"], [79, 12, 1, "", "disable_forward_chunking"], [79, 12, 1, "", "enable_forward_chunking"], [79, 12, 1, "", "forward"], [79, 12, 1, "", "from_pretrained"], [79, 12, 1, "", "fuse_qkv_projections"], [79, 12, 1, "", "load"], [79, 12, 1, "", "prepare_inputs"], [79, 12, 1, "", "set_attn_processor"], [79, 12, 1, "", "unfuse_qkv_projections"]], "tensorrt_llm.models.SpeculativeDecodingMode": [[79, 11, 1, "", "DRAFT_TOKENS_EXTERNAL"], [79, 11, 1, "", "EAGLE"], [79, 11, 1, "", "EXPLICIT_DRAFT_TOKENS"], [79, 11, 1, "", "LOOKAHEAD_DECODING"], [79, 11, 1, "", "MEDUSA"], [79, 11, 1, "", "NONE"], [79, 12, 1, "", "from_arguments"]], "tensorrt_llm.models.WhisperEncoder": [[79, 12, 1, "", "forward"], [79, 12, 1, "", "precompute_relative_attention_bias"], [79, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.plugin": [[80, 10, 1, "", "PluginConfig"]], "tensorrt_llm.plugin.PluginConfig": [[80, 12, 1, "", "to_legacy_setting"]], "tensorrt_llm.quantization": [[81, 10, 1, "", "QuantAlgo"], [81, 10, 1, "", "QuantMode"], [81, 14, 1, "", "quantize_and_export"]], "tensorrt_llm.runtime": [[82, 10, 1, "", "ChatGLMGenerationSession"], [82, 10, 1, "", "EncDecModelRunner"], [82, 10, 1, "", "GenerationSequence"], [82, 10, 1, "", "GenerationSession"], [82, 10, 1, "", "KVCacheManager"], [82, 10, 1, "", "LogitsProcessor"], [82, 10, 1, "", "LogitsProcessorList"], [82, 10, 1, "", "ModelConfig"], [82, 10, 1, "", "ModelRunner"], [82, 10, 1, "", "ModelRunnerCpp"], [82, 10, 1, "", "MultimodalModelRunner"], [82, 10, 1, "", "QWenForCausalLMGenerationSession"], [82, 10, 1, "", "SamplingConfig"], [82, 10, 1, "", "Session"], [82, 10, 1, "", "StoppingCriteria"], [82, 10, 1, "", "StoppingCriteriaList"], [82, 10, 1, "", "TensorInfo"], [82, 14, 1, "", "decode_words_list"]], "tensorrt_llm.runtime.EncDecModelRunner": [[82, 12, 1, "", "encoder_run"], [82, 12, 1, "", "from_engine"], [82, 12, 1, "", "generate"], [82, 12, 1, "", "process_input"]], "tensorrt_llm.runtime.GenerationSequence": [[82, 12, 1, "", "get_batch_idx"], [82, 12, 1, "", "get_seq_idx"]], "tensorrt_llm.runtime.GenerationSession": [[82, 11, 1, "", "batch_size"], [82, 11, 1, "", "buffer_allocated"], [82, 13, 1, "", "context_mem_size"], [82, 13, 1, "", "conv_kernel"], [82, 13, 1, "", "cross_attention"], [82, 11, 1, "", "cuda_graph_mode"], [82, 12, 1, "", "cuda_stream_guard"], [82, 11, 1, "", "debug_mode"], [82, 11, 1, "", "debug_tensors_to_save"], [82, 12, 1, "", "decode"], [82, 12, 1, "", "decode_batch"], [82, 12, 1, "", "decode_regular"], [82, 12, 1, "", "decode_stream"], [82, 11, 1, "", "device"], [82, 13, 1, "", "dtype"], [82, 12, 1, "", "dump_debug_buffers"], [82, 12, 1, "", "early_stop_criteria"], [82, 13, 1, "", "engine_inspector"], [82, 12, 1, "", "filter_medusa_logits"], [82, 12, 1, "", "finalize_decoder"], [82, 12, 1, "", "find_best_medusa_path"], [82, 13, 1, "", "first_layer"], [82, 13, 1, "", "gather_context_logits"], [82, 13, 1, "", "gather_generation_logits"], [82, 13, 1, "", "gemm_allreduce_plugin"], [82, 12, 1, "", "get_next_medusa_tokens"], [82, 12, 1, "", "get_num_heads_kv"], [82, 12, 1, "", "handle_per_step"], [82, 13, 1, "", "has_position_embedding"], [82, 13, 1, "", "has_token_type_embedding"], [82, 13, 1, "", "head_size"], [82, 13, 1, "", "hidden_size"], [82, 13, 1, "", "is_medusa_mode"], [82, 13, 1, "", "is_redrafter_mode"], [82, 13, 1, "", "kv_cache_type"], [82, 13, 1, "", "last_layer"], [82, 12, 1, "", "locate_accepted_draft_tokens"], [82, 11, 1, "", "mapping"], [82, 13, 1, "", "max_draft_tokens"], [82, 13, 1, "", "max_prompt_embedding_table_size"], [82, 12, 1, "", "medusa_decode_and_verify"], [82, 11, 1, "", "medusa_paths"], [82, 11, 1, "", "medusa_position_offsets"], [82, 11, 1, "", "medusa_temperature"], [82, 11, 1, "", "medusa_topks"], [82, 11, 1, "", "medusa_tree_ids"], [82, 12, 1, "", "next_medusa_input_ids"], [82, 11, 1, "", "num_draft_tokens"], [82, 13, 1, "", "num_heads"], [82, 13, 1, "", "num_layers"], [82, 13, 1, "", "num_medusa_heads"], [82, 13, 1, "", "paged_kv_cache"], [82, 13, 1, "", "paged_state"], [82, 12, 1, "", "pp_communicate_final_output_ids"], [82, 12, 1, "", "pp_communicate_new_tokens"], [82, 12, 1, "", "process_logits_including_draft"], [82, 13, 1, "", "profiler"], [82, 13, 1, "", "quant_mode"], [82, 13, 1, "", "remove_input_padding"], [82, 12, 1, "", "reorder_kv_cache_for_beam_search"], [82, 13, 1, "", "rnn_conv_dim_size"], [82, 13, 1, "", "rnn_head_size"], [82, 13, 1, "", "rnn_hidden_size"], [82, 11, 1, "", "runtime"], [82, 12, 1, "", "setup"], [82, 13, 1, "", "state_dtype"], [82, 13, 1, "", "state_size"], [82, 13, 1, "", "tokens_per_block"], [82, 12, 1, "", "update_output_ids_by_offset"], [82, 13, 1, "", "use_gemm_allreduce_plugin"], [82, 13, 1, "", "use_gpt_attention_plugin"], [82, 13, 1, "", "use_kv_cache"], [82, 13, 1, "", "use_lora_plugin"], [82, 13, 1, "", "use_mamba_conv1d_plugin"], [82, 13, 1, "", "vocab_size"]], "tensorrt_llm.runtime.KVCacheManager": [[82, 12, 1, "", "add_sequence"], [82, 12, 1, "", "get_block_offsets"], [82, 12, 1, "", "step"]], "tensorrt_llm.runtime.ModelConfig": [[82, 11, 1, "", "conv_kernel"], [82, 11, 1, "", "cross_attention"], [82, 11, 1, "", "dtype"], [82, 11, 1, "", "gather_context_logits"], [82, 11, 1, "", "gather_generation_logits"], [82, 11, 1, "", "gemm_allreduce_plugin"], [82, 11, 1, "", "gpt_attention_plugin"], [82, 11, 1, "", "gpu_weights_percent"], [82, 11, 1, "", "has_position_embedding"], [82, 11, 1, "", "has_token_type_embedding"], [82, 11, 1, "", "head_size"], [82, 11, 1, "", "hidden_size"], [82, 11, 1, "", "kv_cache_type"], [82, 11, 1, "", "language_adapter_config"], [82, 11, 1, "", "layer_types"], [82, 11, 1, "", "lora_plugin"], [82, 11, 1, "", "lora_target_modules"], [82, 11, 1, "", "mamba_conv1d_plugin"], [82, 11, 1, "", "max_batch_size"], [82, 11, 1, "", "max_beam_width"], [82, 11, 1, "", "max_medusa_tokens"], [82, 11, 1, "", "max_prompt_embedding_table_size"], [82, 11, 1, "", "model_name"], [82, 11, 1, "", "num_heads"], [82, 11, 1, "", "num_kv_heads"], [82, 11, 1, "", "num_kv_heads_per_cross_attn_layer"], [82, 11, 1, "", "num_kv_heads_per_layer"], [82, 11, 1, "", "num_layers"], [82, 11, 1, "", "num_medusa_heads"], [82, 11, 1, "", "paged_state"], [82, 11, 1, "", "quant_mode"], [82, 11, 1, "", "redrafter_draft_len_per_beam"], [82, 11, 1, "", "redrafter_num_beams"], [82, 11, 1, "", "remove_input_padding"], [82, 11, 1, "", "rnn_conv_dim_size"], [82, 11, 1, "", "rnn_head_size"], [82, 11, 1, "", "rnn_hidden_size"], [82, 11, 1, "", "skip_cross_attn_blocks"], [82, 11, 1, "", "skip_cross_kv"], [82, 11, 1, "", "state_dtype"], [82, 11, 1, "", "state_size"], [82, 11, 1, "", "tokens_per_block"], [82, 11, 1, "", "trtllm_modules_to_hf_modules"], [82, 11, 1, "", "vocab_size"]], "tensorrt_llm.runtime.ModelRunner": [[82, 13, 1, "", "dtype"], [82, 12, 1, "", "from_dir"], [82, 12, 1, "", "from_engine"], [82, 13, 1, "", "gather_context_logits"], [82, 13, 1, "", "gather_generation_logits"], [82, 12, 1, "", "generate"], [82, 13, 1, "", "hidden_size"], [82, 13, 1, "", "mapping"], [82, 13, 1, "", "max_prompt_embedding_table_size"], [82, 13, 1, "", "max_sequence_length"], [82, 13, 1, "", "num_heads"], [82, 13, 1, "", "num_layers"], [82, 13, 1, "", "remove_input_padding"], [82, 12, 1, "", "serialize_engine"], [82, 13, 1, "", "use_lora_plugin"], [82, 13, 1, "", "vocab_size"], [82, 13, 1, "", "vocab_size_padded"]], "tensorrt_llm.runtime.ModelRunnerCpp": [[82, 13, 1, "", "dtype"], [82, 12, 1, "", "from_dir"], [82, 13, 1, "", "gather_context_logits"], [82, 13, 1, "", "gather_generation_logits"], [82, 12, 1, "", "generate"], [82, 13, 1, "", "hidden_size"], [82, 13, 1, "", "max_prompt_embedding_table_size"], [82, 13, 1, "", "max_sequence_length"], [82, 13, 1, "", "num_heads"], [82, 13, 1, "", "num_layers"], [82, 13, 1, "", "remove_input_padding"], [82, 13, 1, "", "vocab_size"], [82, 13, 1, "", "vocab_size_padded"]], "tensorrt_llm.runtime.MultimodalModelRunner": [[82, 13, 1, "", "audio_engine_dir"], [82, 13, 1, "", "cpp_e2e"], [82, 13, 1, "", "cpp_llm_only"], [82, 12, 1, "", "generate"], [82, 12, 1, "", "get_audio_features"], [82, 12, 1, "", "get_rope_index"], [82, 12, 1, "", "get_visual_features"], [82, 12, 1, "", "init_audio_encoder"], [82, 12, 1, "", "init_image_encoder"], [82, 12, 1, "", "init_llm"], [82, 12, 1, "", "init_processor"], [82, 12, 1, "", "init_tokenizer"], [82, 13, 1, "", "llm_engine_dir"], [82, 12, 1, "", "load_test_audio"], [82, 12, 1, "", "load_test_data"], [82, 12, 1, "", "prepare_position_ids_for_cogvlm"], [82, 12, 1, "", "preprocess"], [82, 12, 1, "", "ptuning_setup"], [82, 12, 1, "", "ptuning_setup_fuyu"], [82, 12, 1, "", "ptuning_setup_llava_next"], [82, 12, 1, "", "ptuning_setup_phi3"], [82, 12, 1, "", "ptuning_setup_pixtral"], [82, 13, 1, "", "python_e2e"], [82, 12, 1, "", "run"], [82, 12, 1, "", "setup_fake_prompts"], [82, 12, 1, "", "setup_fake_prompts_qwen2vl"], [82, 12, 1, "", "setup_fake_prompts_vila"], [82, 12, 1, "", "setup_inputs"], [82, 12, 1, "", "split_prompt_by_images"], [82, 12, 1, "", "tokenizer_image_token"], [82, 12, 1, "", "video_preprocess"], [82, 13, 1, "", "visual_engine_dir"]], "tensorrt_llm.runtime.QWenForCausalLMGenerationSession": [[82, 12, 1, "", "generate"]], "tensorrt_llm.runtime.SamplingConfig": [[82, 11, 1, "", "bad_words_list"], [82, 11, 1, "", "beam_search_diversity_rate"], [82, 11, 1, "", "early_stopping"], [82, 11, 1, "", "end_id"], [82, 11, 1, "", "frequency_penalty"], [82, 11, 1, "", "length_penalty"], [82, 11, 1, "", "max_attention_window_size"], [82, 11, 1, "", "max_new_tokens"], [82, 11, 1, "", "min_length"], [82, 11, 1, "", "min_p"], [82, 11, 1, "", "no_repeat_ngram_size"], [82, 11, 1, "", "num_beams"], [82, 11, 1, "", "num_return_sequences"], [82, 11, 1, "", "output_cum_log_probs"], [82, 11, 1, "", "output_log_probs"], [82, 11, 1, "", "output_sequence_lengths"], [82, 11, 1, "", "pad_id"], [82, 11, 1, "", "presence_penalty"], [82, 11, 1, "", "random_seed"], [82, 11, 1, "", "repetition_penalty"], [82, 11, 1, "", "return_dict"], [82, 11, 1, "", "sink_token_length"], [82, 11, 1, "", "stop_words_list"], [82, 11, 1, "", "temperature"], [82, 11, 1, "", "top_k"], [82, 11, 1, "", "top_p"], [82, 11, 1, "", "top_p_decay"], [82, 11, 1, "", "top_p_min"], [82, 11, 1, "", "top_p_reset_ids"], [82, 12, 1, "", "update"], [82, 11, 1, "", "use_beam_hyps"]], "tensorrt_llm.runtime.Session": [[82, 13, 1, "", "context"], [82, 13, 1, "", "context_mem_size"], [82, 13, 1, "", "engine"], [82, 12, 1, "", "from_engine"], [82, 12, 1, "", "from_serialized_engine"], [82, 12, 1, "", "infer_shapes"], [82, 12, 1, "", "run"], [82, 13, 1, "", "runtime"], [82, 12, 1, "", "set_shapes"]], "tensorrt_llm.runtime.TensorInfo": [[82, 11, 1, "", "dtype"], [82, 11, 1, "", "name"], [82, 12, 1, "", "numel"], [82, 11, 1, "", "shape"], [82, 12, 1, "", "squeeze"], [82, 12, 1, "", "view"]], "trtllm-serve-disaggregated": [[26, 16, 1, "cmdoption-trtllm-serve-disaggregated-c", "--config_file"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated-r", "--request_timeout"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated-t", "--server_start_timeout"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated-c", "-c"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated-r", "-r"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated-t", "-t"]], "trtllm-serve-disaggregated_mpi_worker": [[26, 16, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", "--config_file"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", "--log_level"], [26, 16, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", "-c"]], "trtllm-serve-serve": [[26, 16, 1, "cmdoption-trtllm-serve-serve-backend", "--backend"], [26, 16, 1, "cmdoption-trtllm-serve-serve-cluster_size", "--cluster_size"], [26, 16, 1, "cmdoption-trtllm-serve-serve-ep_size", "--ep_size"], [26, 16, 1, "cmdoption-trtllm-serve-serve-extra_llm_api_options", "--extra_llm_api_options"], [26, 16, 1, "cmdoption-trtllm-serve-serve-gpus_per_node", "--gpus_per_node"], [26, 16, 1, "cmdoption-trtllm-serve-serve-host", "--host"], [26, 16, 1, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", "--kv_cache_free_gpu_memory_fraction"], [26, 16, 1, "cmdoption-trtllm-serve-serve-log_level", "--log_level"], [26, 16, 1, "cmdoption-trtllm-serve-serve-max_batch_size", "--max_batch_size"], [26, 16, 1, "cmdoption-trtllm-serve-serve-max_beam_width", "--max_beam_width"], [26, 16, 1, "cmdoption-trtllm-serve-serve-max_num_tokens", "--max_num_tokens"], [26, 16, 1, "cmdoption-trtllm-serve-serve-max_seq_len", "--max_seq_len"], [26, 16, 1, "cmdoption-trtllm-serve-serve-num_postprocess_workers", "--num_postprocess_workers"], [26, 16, 1, "cmdoption-trtllm-serve-serve-port", "--port"], [26, 16, 1, "cmdoption-trtllm-serve-serve-pp_size", "--pp_size"], [26, 16, 1, "cmdoption-trtllm-serve-serve-reasoning_parser", "--reasoning_parser"], [26, 16, 1, "cmdoption-trtllm-serve-serve-tokenizer", "--tokenizer"], [26, 16, 1, "cmdoption-trtllm-serve-serve-tp_size", "--tp_size"], [26, 16, 1, "cmdoption-trtllm-serve-serve-trust_remote_code", "--trust_remote_code"], [26, 16, 1, "cmdoption-trtllm-serve-serve-arg-MODEL", "MODEL"]]}, "objnames": {"0": ["c", "macro", "C macro"], "1": ["cpp", "type", "C++ type"], "2": ["cpp", "class", "C++ class"], "3": ["cpp", "function", "C++ function"], "4": ["cpp", "functionParam", "C++ function parameter"], "5": ["cpp", "member", "C++ member"], "6": ["cpp", "enum", "C++ enum"], "7": ["cpp", "enumerator", "C++ enumerator"], "8": ["cpp", "templateParam", "C++ template parameter"], "9": ["py", "module", "Python module"], "10": ["py", "class", "Python class"], "11": ["py", "attribute", "Python attribute"], "12": ["py", "method", "Python method"], "13": ["py", "property", "Python property"], "14": ["py", "function", "Python function"], "15": ["py", "pydantic_field", "Python field"], "16": ["std", "cmdoption", "program option"]}, "objtypes": {"0": "c:macro", "1": "cpp:type", "2": "cpp:class", "3": "cpp:function", "4": "cpp:functionParam", "5": "cpp:member", "6": "cpp:enum", "7": "cpp:enumerator", "8": "cpp:templateParam", "9": "py:module", "10": "py:class", "11": "py:attribute", "12": "py:method", "13": "py:property", "14": "py:function", "15": "py:pydantic_field", "16": "std:cmdoption"}, "terms": {"": [0, 1, 2, 3, 4, 6, 7, 10, 12, 13, 14, 15, 16, 17, 18, 19, 21, 22, 24, 25, 27, 40, 41, 45, 46, 47, 53, 60, 64, 65, 66, 68, 70, 72, 73, 74, 75, 77, 78, 79, 82, 83, 84, 85, 87, 88, 90, 91, 92, 93], "0": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 17, 18, 20, 21, 23, 24, 25, 26, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 51, 52, 53, 54, 55, 57, 59, 60, 61, 62, 64, 65, 66, 67, 68, 69, 70, 74, 75, 76, 77, 78, 79, 82, 83, 84, 86, 87, 89, 90, 94], "00": [14, 24, 50, 51, 52, 68, 69, 70, 87], "000": [18, 68], "0000": [68, 70], "0007503032684326172": 26, "001": 46, "0012": 68, "0017": 69, "003": 69, "0047": 87, "005": 69, "0070": 87, "0071": 87, "0096": 87, "00978": 85, "01": [23, 24, 50, 51, 52, 68, 69, 84, 88], "014": 21, "0158": 70, "016": 69, "0162": 72, "0165": 74, "017": 69, "02": [69, 88], "021": 69, "022": 69, "0235": 87, "0260": 87, "0273": 87, "028": 69, "0294": 87, "03": [74, 87, 88], "032": 24, "0339": 69, "03762": 77, "03961": 4, "04": [61, 62, 69, 86, 88, 89], "043": 69, "0449": 87, "0461": 18, "0463": 69, "05": [69, 77, 78, 79, 87, 88], "05100": 77, "0523": 87, "055": 69, "0554": 70, "0560": 87, "0563": 69, "06": [24, 68, 69, 77, 78], "0630": 87, "0669": 18, "068": 69, "0682": 87, "0689e": 68, "07": [23, 24, 69, 88], "0704": 70, "0713": 87, "0723": 87, "0732": 87, "0758": 18, "0772": 18, "0776": 87, "08": [24, 69, 74], "0804": 87, "082": 69, "0838": 69, "0881": 75, "089": 69, "09": [24, 87], "0903": 87, "0910": 87, "092": 69, "09353": 9, "0964": 69, "09685": 9, "097": 69, "09f": [0, 1], "0b": 2, "0e": 6, "0f": [0, 6, 65], "0rc1": 68, "0u": 1, "0x": 20, "0x0000000000000000": 88, "1": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 20, 21, 22, 23, 24, 25, 26, 29, 31, 32, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 53, 54, 55, 57, 59, 61, 62, 64, 65, 67, 68, 70, 71, 72, 74, 76, 77, 78, 79, 81, 82, 83, 86, 87, 89, 93], "10": [0, 8, 9, 10, 18, 23, 24, 26, 32, 34, 39, 46, 54, 59, 62, 65, 68, 69, 70, 72, 75, 77, 84, 86, 87], "100": [0, 8, 18, 26, 34, 51, 67, 68, 70, 83], "1000": [0, 67, 68, 69, 70], "10000": [77, 78, 79], "1003": 88, "101": 8, "101230": 46, "101978": 69, "102": [8, 20], "1024": [1, 6, 13, 18, 21, 23, 25, 32, 39, 46, 49, 65, 68, 69, 70, 74, 77, 78, 87], "103": 8, "104": 88, "10438": 85, "1045": 87, "1047": 68, "1050": 87, "1051": 70, "1059": 68, "106563": 69, "1072": 87, "107501": 69, "10764": 48, "10774": 0, "1079": 17, "108": 69, "1082": 87, "10858": 32, "10b": [64, 77, 88], "10m": 20, "11": [0, 9, 10, 18, 21, 23, 59, 68, 69, 72, 77, 87], "11023": 68, "110804": 69, "110b": 88, "111": [20, 24], "111302": 69, "111618": 69, "111668": 69, "1118": 88, "1123": 88, "1134": 84, "1135": 87, "1141": 87, "1148": 88, "11489": 18, "11490": 68, "1151": 18, "115716": 69, "1160": [26, 33], "117": 69, "1178": 68, "1181": 88, "1183": 88, "119": 68, "11943": 68, "11947": 32, "1196": 18, "11b": [86, 88], "12": [0, 9, 13, 20, 24, 32, 59, 61, 62, 68, 69, 72, 74, 77, 87], "1207": 48, "1212": 87, "121847": 68, "1219": 18, "122": 68, "1225": 77, "12288": 68, "123": [26, 34, 35], "1234": [65, 79], "1239": 88, "1242": 88, "1248": 88, "125": 68, "1252": [17, 68], "1256": 88, "125m": [10, 13], "126": 68, "1267": 88, "127": 77, "1272": 87, "128": [0, 1, 5, 8, 9, 11, 14, 18, 19, 20, 21, 22, 23, 24, 26, 32, 34, 35, 44, 51, 65, 68, 69, 88], "1284": 88, "1287": 72, "1290": 87, "1291504": 70, "1293": 17, "12945": 18, "129498": 18, "13": [5, 9, 22, 59, 68, 69, 70, 77, 87], "1300": 40, "13044": 48, "131072": [68, 70], "13195": 68, "132": [68, 69], "1323": 88, "1328": 88, "1329": 88, "133": 88, "13368": 68, "1337": 88, "1341": 18, "1343": 88, "1344": 88, "13525": 68, "13598": 68, "1363": 48, "137": 68, "1378": 87, "139": 69, "1392": 88, "13b": 20, "14": [9, 13, 23, 59, 68, 69, 72, 74, 75, 87], "140g": 17, "141": 21, "1418": 68, "141gb": [19, 69], "1424": 88, "1436": [18, 88], "1437": 87, "144": 72, "1446": 88, "1447": 88, "14480": 68, "1449": 88, "145": [74, 75], "1459": 87, "146": [74, 75], "1467": 88, "147": [70, 72, 74, 75], "1480": 88, "1486": 88, "149": [87, 88], "15": [9, 24, 59, 68, 69, 75, 77, 87], "150": 67, "1500": 69, "15043": 32, "1514": 88, "1529": 88, "1534": 88, "1535": 88, "1536": 18, "1537": 88, "1539": 88, "154": 24, "1552": 88, "1556": 87, "15585": 68, "1562": 88, "1564": [70, 74, 75], "158": 18, "1583": 88, "1584": 18, "1585": 70, "15889": 48, "1589": 88, "1590": 88, "1597": 72, "16": [0, 5, 9, 10, 14, 18, 20, 23, 24, 26, 29, 31, 50, 51, 52, 59, 60, 68, 69, 70, 71, 77, 78, 79, 84, 85, 87], "160": 88, "1607": 68, "161": [26, 33, 68], "1625": 72, "1626": 88, "163": 19, "1637": 88, "16384": [72, 74], "164": 24, "1642": 88, "1650": 88, "1660": 88, "1669": 88, "167": [68, 69], "1672": 87, "1674": 88, "1675": 88, "1676": 88, "168": 24, "16e": 86, "16x": 84, "17": [0, 2, 9, 18, 59, 68, 69, 74, 87, 89], "1706": 77, "1721": 87, "1723": 88, "17233": 18, "173": 24, "1732": 88, "17323": 85, "1738": 88, "174": 69, "1741966075": 83, "1742": 88, "17453": 25, "17453v3": 1, "175": 69, "175b": 21, "176": 68, "176064": 18, "1762": 88, "1799": 88, "17b": 86, "18": [2, 9, 59, 66, 68, 69, 87], "180": [24, 84], "180000000": 0, "180b": [23, 68], "1811": 48, "1815": 88, "181540": 18, "182": 69, "1822": 32, "183": 69, "1834": 88, "184": 69, "185": [20, 68], "1851": 88, "18527": 32, "18533": 48, "18563": 68, "1861": 75, "1866": 75, "1885": 70, "1886": 88, "1889": 48, "1897": 88, "19": [2, 18, 59, 69, 75, 87], "1900": 48, "1909": 88, "191": 69, "192": 19, "1921": 18, "1926": 88, "1937": 88, "1939": 88, "1944": 74, "1953": 88, "1959": 68, "198": 24, "1985": 88, "1987": 88, "1993": 87, "1999": 88, "1_405b": 14, "1_70b": 14, "1b": [26, 29, 31, 34, 36, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48, 53, 54, 55, 57, 61, 62, 64, 83, 89], "1d": [5, 77, 82], "1e": [13, 77, 78, 79], "1e20f": 1, "1g": 87, "1gb": 2, "1k": [18, 24], "1m": 75, "1st": [20, 77, 84], "1u": [0, 1], "1x": 24, "1xh200": 19, "1ytic": 88, "2": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 13, 14, 17, 19, 20, 21, 23, 24, 26, 38, 39, 43, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 59, 61, 62, 64, 65, 68, 69, 71, 72, 74, 75, 77, 79, 82, 85, 86, 87, 93], "20": [1, 6, 10, 11, 26, 55, 57, 68, 69, 70, 74, 77, 82, 87], "200": [21, 82], "2000": 69, "20000": 69, "2017": 74, "2018": 88, "2023": [19, 87], "2024": 24, "2025": [18, 24, 68], "2028": 88, "203": 69, "2033": 75, "2039": 88, "204": [24, 69], "2040": 88, "2044": [74, 75], "2045": 74, "2048": [13, 18, 19, 21, 22, 25, 44, 65, 68, 69, 70, 72, 73, 74, 75, 79, 82, 87, 88], "2056": 88, "206": 69, "20627": 32, "20685": 68, "2079": 87, "208": 69, "2081": [72, 74, 88], "2087": 88, "2089": 69, "209": 69, "20b": 88, "21": [10, 23, 24, 69, 74, 87, 88], "2101": 4, "2102": 69, "2106": 9, "2107": [48, 87], "210g": 17, "211": 24, "2113": 88, "2135": 88, "21367": 48, "2152": 88, "2158": 69, "2168": 18, "2169": 88, "21747": 68, "2176": 69, "21764": 68, "2182": 88, "2191": 88, "22": [28, 69, 77, 87], "22000": 69, "22056": 68, "221": 68, "2210": 85, "2211": [77, 85], "2219": 88, "22213": 68, "2225": 87, "2232": 88, "224": 78, "2243": 88, "2263": 88, "227": 22, "2288": 88, "2294": 88, "23": [68, 69, 87, 88], "2305": 87, "2306": 85, "2309": [1, 25], "232": 22, "2337": 48, "2352": 88, "2357": 88, "236": 24, "2366": 88, "2370": 88, "2373": 88, "2379": 88, "2388": 88, "239": 24, "2397": 68, "24": [0, 61, 62, 69, 87, 88, 89], "240": 69, "2401": 0, "2402": 9, "24189": 69, "2419": 88, "242": 69, "2425": 88, "2439": 88, "245": 24, "2458": 88, "2461": 74, "2466": 74, "2473": 88, "2474": [72, 74], "2484": 88, "2485": 88, "2487": 69, "249": 24, "25": [22, 24, 68, 69, 86, 88], "250": [18, 24], "2500": 69, "25032": 68, "253": [24, 69], "2552": 88, "256": [1, 18, 19, 22, 54, 65, 68, 69, 77, 87, 88], "25603": 68, "2573": 88, "2581": [72, 74], "2590780": 68, "259840": 84, "26": [68, 69, 72, 83], "260": 69, "2602": 32, "2628": [74, 75], "263": [19, 32, 48], "2640": 75, "2649": 87, "2671": 18, "2677": 88, "26778": 68, "2679": 72, "2685": 88, "2688": 48, "2691": 88, "27": [69, 88], "270": 69, "2712": 88, "274": [18, 88], "2742": 70, "275": 88, "27556": 48, "276": 69, "278": [32, 48, 69], "2782": 88, "2787": 88, "2796": 88, "28": [24, 68, 69, 87], "2820": 87, "2826700": 18, "28390": 68, "287113": 68, "288": 88, "29": [69, 84], "292": 69, "2939": 87, "294": 69, "297": 32, "29889": 48, "29892": 32, "299": [24, 68], "29962": 32, "2998": 87, "2b": [17, 59, 68], "2d": [10, 77, 78, 85], "2k": [18, 24], "2m": 75, "2nd": 77, "2u": 1, "2x": [20, 21], "3": [0, 1, 3, 5, 7, 8, 9, 15, 19, 20, 21, 23, 24, 39, 41, 43, 47, 49, 53, 54, 59, 61, 62, 64, 65, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 82, 83, 87, 88, 89, 90], "30": [0, 10, 18, 24, 65, 69, 70, 72, 75, 77, 84], "300": [22, 68], "3000": [68, 69], "30000": 69, "30065": 68, "3019": 68, "3021": 18, "3022": 68, "303": 21, "3031": 74, "304": [32, 48], "3040": [70, 74, 75], "306": 32, "3072": 18, "30990": 68, "30b": 23, "30x": 23, "31": [69, 70, 74, 75], "311": 69, "3132": 68, "315": [24, 69], "318": 69, "32": [1, 5, 8, 18, 20, 21, 25, 32, 48, 65, 68, 69, 70, 77, 78, 79, 82, 83, 84, 85, 87, 88, 89], "3201": 70, "321": 68, "322": [32, 48], "3276": [70, 74, 75], "32768": 77, "3291": 87, "32b": 88, "32k": 88, "32x": 23, "33": [69, 87], "332": 69, "3328": 87, "3338": 70, "338": [24, 32, 48], "3389": 72, "34": [18, 69], "340": [24, 69], "341": 21, "3442": 87, "3445": 87, "3452": 87, "3476": 18, "349": 21, "34b": 88, "35": [0, 65, 69], "351": 69, "3555": 87, "35611": 18, "357": 69, "36": [24, 69, 71, 72], "3671": 68, "368": 24, "37": 68, "370": 69, "371": 69, "374": 69, "375": 69, "3763": 24, "379": 69, "38": [68, 69], "384": [18, 69], "3863": 69, "387": 69, "387b12598a9e": 68, "3885": 18, "3887": 87, "39": [24, 69], "3914": 69, "3936": 68, "3977": 87, "399": 69, "3_1": 86, "3_3": 86, "3b": [30, 35, 56], "3d": [5, 77, 82], "3rd": 77, "3u": 1, "3x": [23, 24], "4": [0, 1, 2, 7, 8, 9, 10, 14, 17, 21, 23, 24, 26, 32, 39, 44, 48, 49, 50, 51, 52, 59, 65, 68, 69, 70, 72, 73, 74, 75, 76, 77, 79, 82, 83, 84, 85, 86, 87, 88], "40": [6, 69, 72, 77, 88], "403": 88, "405": 48, "405b": [68, 71], "4060": 84, "4066": 32, "408": 69, "4089": 75, "4096": [19, 32, 68, 69, 72, 77, 78, 82], "40b": 23, "40gb": 25, "40x": 23, "41": 69, "41020": 68, "411": 68, "4115": 24, "4117e": 68, "4133": 75, "41375": 68, "414": 18, "41607": 68, "4168": 18, "4192": 87, "42": [47, 68, 69], "4203099703668305365": 46, "4224": 69, "4248": 72, "4265": 68, "427": [48, 68, 69], "4280": 24, "43": [69, 83, 84], "433": 69, "437": 69, "438": 69, "44": [69, 84], "4408": 32, "442": 69, "4439": 68, "4451": 18, "4456": 69, "447": 69, "448": 69, "449": 88, "4493": [18, 74, 75], "4497": 69, "44x": 23, "45": [8, 69, 86, 88], "450": 69, "45000000000": 8, "453": 69, "4566": 69, "459": 69, "46": 23, "462": 69, "463": 69, "4653": 32, "4656": 69, "466": 69, "4667": 69, "47": [23, 72], "4701": 68, "471": 69, "472": 32, "475": 69, "477": 69, "478": 88, "47x": 23, "48": [69, 72, 84, 88], "481": [20, 69], "482": 88, "488": 69, "49": [69, 72], "49152": 18, "495": 69, "4963": 68, "49b": 86, "4b": 88, "4bit": 19, "4u": 1, "4x": [19, 20, 21], "5": [0, 1, 8, 9, 10, 11, 13, 19, 20, 21, 23, 24, 30, 35, 39, 40, 46, 49, 56, 64, 65, 68, 69, 74, 77, 79, 82, 86, 87, 88], "50": [0, 23, 40, 65, 68, 69, 88], "500": [24, 69], "5000": 69, "500000": 79, "5001": 48, "5007": 32, "500m": 23, "50272": 13, "505143404006958": 26, "5064": 69, "5073": 87, "51": 69, "512": [1, 9, 11, 21, 22, 65, 68, 69, 72, 74, 79], "5120": 18, "512mb": 2, "514": 69, "518": [32, 69], "51b": [86, 88], "51x": 23, "52269": 69, "524": 69, "525": 69, "526": [48, 69, 88], "52667": 69, "529": 69, "5299": 72, "53": [68, 74, 75], "5305": 72, "531": 69, "54": [23, 69], "540": 68, "543": 69, "544": 69, "5496": 72, "5497": 69, "55": [23, 68, 69], "5500": 69, "5510": 68, "5514": 68, "5530": 69, "554": 69, "557": 69, "559": 69, "56": [23, 69], "560": 19, "562": [9, 11], "56401920000": 26, "565": 69, "567": 69, "568": [68, 69], "57": [68, 69], "571": 69, "572": 69, "5739": 18, "5742": [72, 74], "579": 69, "58": [24, 69, 74], "580": 69, "5821": 69, "5830": 87, "5874": 87, "5877": 72, "5879": 87, "588": 69, "58x": 24, "59": 68, "590": [32, 69], "5918": 87, "5942": 18, "5957": 87, "5976": 72, "598": 69, "5980": 72, "5b": 88, "5th": 77, "5u": 1, "5x": [20, 23, 24], "6": [0, 1, 6, 8, 9, 10, 21, 23, 24, 26, 39, 49, 65, 69, 77, 82, 86, 87, 88], "60": [0, 69], "600": 27, "6000": 68, "602": 69, "6049": 72, "6059": 68, "6064": 87, "608": 69, "61": 69, "610": 69, "6100": 18, "6157": 87, "618": 69, "62": [24, 69, 74], "6255": 87, "626": 32, "6299": 87, "63": [39, 49, 60, 68, 69, 74, 79, 84], "630": 69, "63266": 70, "63307": 70, "63308": 70, "63331": 70, "63374": 70, "634": 69, "63456": 70, "6345624": 70, "6372": 72, "639": 88, "64": [0, 1, 5, 6, 13, 18, 20, 21, 25, 30, 35, 53, 56, 69, 74, 77, 78, 79, 84, 88], "640": [19, 69], "6452": 75, "6475": 74, "649": 88, "64x": 24, "65": [62, 69], "65024": 87, "6523": 75, "653": 69, "654": 21, "6550": 72, "6554": 74, "656": 69, "657": 69, "659": 69, "6591": 68, "66": [24, 69], "661": 69, "6628": [74, 75], "6678": 84, "6684": 75, "6695": 84, "67": [23, 24, 69], "6701": 18, "671": 18, "67108864": 60, "673": 88, "675": 68, "6753e": 68, "6769": 74, "679": 20, "68": [23, 24, 69, 75], "682": 69, "6825": 68, "683": 69, "684": 24, "685": 69, "6852": [72, 74], "686": 69, "6862": 68, "6890": 87, "69": [23, 24, 69, 75, 83], "6925": 68, "6938": 32, "695": 88, "696": 69, "6975": 72, "6976": [70, 74, 75], "698": 69, "6a": 19, "6b": [20, 68, 77, 88], "6x": 21, "7": [0, 1, 8, 9, 19, 20, 23, 24, 39, 49, 59, 60, 61, 62, 68, 69, 70, 77, 82, 87], "70": [0, 23, 75, 84], "700": 27, "7000": 68, "701": 88, "7031": 72, "704": 69, "705": 88, "706": 69, "7063": 68, "707": 69, "7072": 69, "709": 68, "7090": 87, "70b": [5, 17, 21, 23, 49, 70, 72, 73, 74, 75, 76, 86, 88], "70g": 17, "71": [24, 68, 69], "711": 69, "712": 69, "7134": 87, "7136": 70, "714": 69, "7144": 87, "7168": 24, "717": 69, "7187": 69, "7188": 18, "72": [69, 71], "722": 69, "727": 69, "72b": [86, 88], "73": [24, 69], "732": 69, "734": 69, "736": 69, "737": 69, "7382": 69, "739": 88, "74": [24, 69], "741": [69, 88], "742": 69, "745": 69, "7456": 18, "74561": 18, "747": 69, "7480": 70, "75": [23, 68, 88], "750": [21, 69], "7502": 70, "7520": 18, "755": 27, "7584": 18, "75903": 69, "76": 69, "7607": 74, "7621": 69, "7638": [70, 74, 75], "767": 69, "768": [13, 78], "77": 69, "772": 69, "7743": 70, "7770": 70, "78": [24, 69, 72], "780": 68, "7842": 72, "78509": 69, "7876": 74, "79": [68, 84], "7900": 87, "7933": 74, "794": [69, 88], "7949": 87, "7977": 72, "7a": 19, "7b": [9, 10, 11, 23, 26, 39, 49, 68, 69, 83, 86, 88], "7x": [20, 24], "8": [0, 1, 5, 8, 9, 13, 14, 17, 18, 19, 21, 22, 23, 24, 25, 26, 32, 33, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 50, 51, 52, 54, 59, 61, 62, 65, 68, 69, 70, 71, 72, 76, 77, 78, 79, 83, 84, 85, 87], "80": [0, 6, 21, 24, 60, 69, 88], "800": [19, 69, 88], "8000": [26, 29, 30, 31, 33, 34, 35, 55, 56, 57, 83], "8002": 68, "8005": 69, "803": 19, "8048": 68, "80gb": [20, 23, 25, 69, 70, 72, 73], "81": [24, 69, 72], "810": 69, "8149": 87, "8179": 87, "819": 21, "8192": [25, 65, 68, 69, 70, 74, 77, 78, 87, 88], "82": [24, 69, 72], "820": 68, "8212": 1, "8218": 87, "822": 69, "8225": 72, "825": 88, "8259": 68, "83": 69, "8307": 75, "8351": 68, "838": 69, "84": [24, 69], "840": 69, "841": 69, "8441": 68, "85": [18, 23, 68, 69, 88], "850": 69, "851": 69, "854": 69, "86": [60, 69], "863": 68, "866": 69, "867": 69, "8672": 87, "87": [23, 69], "8779": 87, "88": [69, 72, 75], "8804": 70, "8828": 87, "8841": 72, "89": [23, 24, 60, 69, 86], "893": 69, "8932": 68, "8958": 75, "896": [48, 69], "8a": 22, "8b": [41, 49, 64, 68, 83, 86, 89], "8bit": 20, "8tb": 21, "8x7b": [4, 68, 86, 88], "8xb200": 24, "8xh100": 22, "8xh200": 19, "9": [0, 1, 9, 10, 17, 20, 24, 39, 49, 54, 59, 69, 72, 77, 87], "90": [0, 18, 60, 65, 68, 69, 70, 72, 76, 84], "9007": 18, "9028": 87, "907": 20, "9087": 75, "91": 69, "910": 69, "9101": 69, "911": 69, "9115": 75, "912656": 18, "913": 69, "9184": 72, "92": [24, 69], "920": 69, "9203": 72, "9214": 69, "924": 13, "925": 69, "9274": 70, "93": 69, "935": 88, "9353e": 70, "9379": 18, "94": 69, "94022": 69, "941": [19, 22], "943": 48, "944": 69, "946": 19, "947": 69, "9494": 74, "95": [26, 33, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 54, 61, 62, 69, 70, 76, 83], "9521": 87, "953": 69, "9537": 72, "956": 69, "957": 69, "96": [19, 24, 69, 72, 88], "960": 19, "961": 69, "9623": 74, "963": 69, "9639": 69, "96583": 69, "967": 88, "9692": 87, "97": [68, 69, 72], "970": 69, "98": 69, "983": 88, "987": 88, "99": [8, 24, 27, 69], "990": 69, "991": 69, "992": 88, "9928": 75, "9938": 18, "9982": [74, 75], "9x": [21, 22], "A": [0, 1, 2, 3, 5, 6, 9, 10, 13, 14, 17, 18, 23, 24, 47, 50, 51, 52, 53, 65, 67, 68, 69, 77, 82, 88, 90, 92], "AND": 77, "And": [10, 17, 77, 78, 84], "As": [4, 5, 7, 9, 10, 14, 16, 32, 72, 75, 76, 77, 84, 85, 87, 92, 93], "At": [12, 53, 72, 78, 84], "But": [5, 66], "By": [0, 1, 2, 6, 10, 24, 32, 60, 68, 72, 75, 77, 87, 92], "For": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 32, 36, 47, 50, 51, 52, 58, 60, 64, 68, 69, 70, 71, 72, 74, 75, 76, 77, 82, 83, 84, 87, 88, 90, 91, 92, 93, 94], "If": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 17, 23, 25, 26, 27, 28, 60, 61, 62, 64, 65, 66, 68, 70, 71, 72, 74, 75, 76, 77, 79, 82, 84, 86, 87, 88, 90, 92, 93, 94], "In": [0, 1, 2, 7, 10, 14, 15, 17, 18, 20, 23, 24, 28, 32, 49, 53, 59, 60, 68, 69, 70, 71, 72, 74, 75, 77, 83, 84, 85, 86, 87, 88, 92, 93, 94], "It": [0, 1, 3, 5, 6, 7, 9, 10, 12, 14, 15, 16, 18, 19, 22, 23, 24, 25, 32, 46, 53, 60, 65, 66, 68, 69, 72, 73, 74, 75, 76, 77, 83, 85, 87, 90, 91, 92, 94], "Its": [5, 77, 92], "NOT": 77, "No": [0, 2, 8, 53, 68, 70], "Not": [1, 23], "ON": [68, 72, 74, 75], "OR": 77, "Of": [24, 88], "On": [5, 8, 60, 62, 67, 71, 75, 77, 88], "One": [2, 13, 14, 74, 77, 87, 91], "Or": [77, 82, 89], "That": [3, 5, 6, 8, 14, 66, 72, 77], "The": [0, 1, 2, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 23, 24, 25, 26, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 47, 48, 49, 50, 51, 52, 53, 54, 59, 60, 61, 62, 64, 65, 67, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 79, 80, 82, 83, 84, 86, 87, 88, 89, 90, 91, 92, 93, 94], "Then": [9, 17, 26, 27, 68, 70, 77, 90, 93], "There": [2, 5, 6, 7, 8, 9, 13, 17, 21, 24, 32, 60, 62, 64, 77, 80, 84, 85, 87, 88, 91, 92, 93, 94], "These": [2, 10, 17, 19, 21, 22, 24, 32, 68, 70, 71, 78, 80, 83, 88], "To": [2, 3, 5, 8, 9, 10, 11, 14, 15, 16, 17, 18, 21, 24, 60, 64, 65, 66, 67, 68, 69, 72, 74, 75, 76, 77, 83, 84, 85, 88, 89, 90, 92, 93, 94], "Will": 0, "With": [5, 6, 10, 14, 27, 32, 44, 59, 68], "_": [0, 3, 15, 80], "__all__": 90, "__call__": 47, "__init__": [7, 12, 14, 15, 47, 65, 68, 87, 88, 90, 92, 94], "__main__": [36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 61, 62, 64, 70, 72, 75, 76, 83, 88, 89, 90], "__name__": [36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 61, 62, 70, 72, 75, 76, 83, 88, 89, 90], "__post_init__": 88, "__repr__": 88, "_capac": 1, "_context_logits_auto_en": 65, "_cpp_gen": 3, "_create_tensor": 14, "_explicitly_disable_gemm_plugin": 80, "_generation_logits_auto_en": 65, "_handl": 1, "_mark_output": 87, "_note": 5, "_path": 18, "_postproc_param": 65, "_postprocess_result": 65, "_return_log_prob": 65, "_run": 87, "_runtim": 82, "_str_to_trt_dtype_dict": 77, "_torch": [46, 68, 88, 89, 90, 91, 92], "_unsign": 1, "_util": 77, "a10": 25, "a100": [6, 17, 25], "a10g": 25, "a2": 88, "a30": 25, "a40": 25, "a8": 85, "a_": 77, "a_1": 77, "a_2": 77, "a_n": 77, "a_sf": 77, "aarch64": 86, "ab": [9, 25, 77, 85], "abbrevi": 26, "abi": [60, 88], "abil": [66, 68], "abl": [5, 20, 24, 62, 68, 74, 77, 88], "abnorm": 88, "abort": [65, 88], "about": [0, 1, 3, 17, 18, 19, 20, 22, 23, 46, 53, 54, 59, 68, 70, 72, 73, 75, 77, 83, 84, 87, 88], "abov": [2, 9, 14, 17, 18, 23, 32, 60, 68, 69, 70, 72, 75, 84], "absenc": 6, "absorb": 24, "abstract": [75, 78], "ac": 88, "acc": 77, "acceler": [5, 10, 20, 21, 22, 23, 25, 66], "accept": [0, 1, 10, 18, 32, 41, 42, 43, 44, 45, 60, 65, 70, 72, 77, 82, 83, 86, 88, 92], "accept_length": 82, "acceptancer": 0, "acceptancethreshold": 0, "acceptedlen": 1, "acceptedlengthscumsum": 1, "acceptedpath": 1, "acceptedpathid": 1, "acceptedtoken": 1, "acceptedtokenslen": 1, "access": [3, 28, 40, 68, 70, 77, 83, 88], "accessor": 1, "accommod": [4, 91, 93], "accomplish": 71, "accord": [5, 15, 54, 77, 78, 92], "accordingli": 15, "account": [14, 18, 27, 50, 51, 52, 60], "accumul": [0, 5, 6, 25, 65, 77, 82, 83], "accur": [19, 40, 68, 70, 88], "accuraci": [19, 24, 25, 72, 76, 77, 85, 88], "achiev": [2, 10, 18, 19, 23, 24, 60, 69, 70, 72, 74, 76, 90], "across": [2, 4, 5, 6, 7, 14, 15, 21, 24, 26, 69, 71, 72, 74, 75, 77, 82], "act": 24, "act_fn": 78, "act_typ": [14, 77], "action": 49, "activ": [0, 1, 5, 7, 14, 18, 19, 20, 23, 24, 25, 71, 77, 85, 86, 88, 94], "activation_scaling_factor": 13, "activationtyp": [14, 77], "active_request": 94, "actual": [7, 10, 18, 23, 24, 25, 72, 74, 75, 76, 88, 93], "ad": [1, 5, 6, 7, 8, 10, 11, 17, 28, 59, 67, 71, 74, 75, 77, 79, 82, 88, 89, 91], "ada": [5, 23, 54, 60, 66, 72, 86, 88], "adalayernorm": 78, "adalayernormcontinu": 78, "adalayernormzero": 78, "adalayernormzerosingl": 78, "adapt": [0, 9, 36, 37, 65, 77, 78, 88, 90], "adapter_s": 9, "adapters": 1, "add": [1, 3, 5, 7, 9, 12, 13, 14, 17, 27, 28, 47, 49, 60, 64, 65, 68, 70, 72, 75, 77, 82, 87, 88, 90, 93], "add_activ": 14, "add_argu": 49, "add_bias_linear": 79, "add_generation_prompt": 24, "add_input": 77, "add_output": 77, "add_padding_request": 93, "add_qkv_bia": 79, "add_rmsnorm": 24, "add_sequ": 82, "add_special_token": [24, 65, 82, 88], "addcumlogprob": 88, "added_kv_proj_dim": 78, "added_proj_bia": 78, "addit": [0, 5, 6, 9, 10, 14, 17, 21, 26, 32, 40, 60, 65, 68, 69, 71, 72, 74, 77, 78, 85, 86, 87, 88, 92, 93], "addition": [2, 68, 70, 72, 75, 90, 92], "additional_model_output": 65, "additional_opt": 52, "additionalmodeloutput": [0, 3, 65], "additionaloutput": [0, 3], "address": [1, 15, 18, 23, 24, 64, 75, 84, 88], "addresswiths": 1, "adequ": 78, "adher": 40, "adjust": [50, 65, 68, 70, 84, 94], "admin": 62, "adopt": [6, 17], "advanc": [10, 14, 22, 24, 25, 38, 41, 42, 44, 45, 60, 65, 77, 88, 92], "advantag": [6, 66], "advers": [19, 25], "advertis": 68, "advis": 2, "affect": [17, 18, 25, 70, 72, 74, 75, 84], "affin": 78, "after": [0, 1, 3, 5, 7, 8, 9, 10, 14, 15, 24, 25, 26, 27, 46, 49, 60, 64, 65, 68, 72, 74, 75, 76, 77, 78, 80, 83, 84, 88, 92, 94], "again": [14, 70, 72, 75, 87], "against": [60, 68], "agent": 21, "aggress": [13, 72, 76], "agre": [64, 83], "ahead": [0, 5, 10], "ai": [18, 20, 24, 26, 33, 36, 38, 39, 41, 42, 43, 44, 45, 49, 54, 61, 62, 66, 67, 70, 76, 77, 83, 86, 88, 89], "aidc": 88, "aim": [4, 13, 18, 24, 66, 68, 70, 72, 88], "ainsli": 19, "air": 88, "aka": 77, "akhoroshev": 88, "al": 19, "albeit": 10, "alessionetti": 88, "algorithm": [0, 5, 6, 10, 13, 14, 17, 23, 24, 65, 68, 72, 77, 88], "alia": [78, 79], "alibi": 77, "alibi_bias_max": [77, 78], "alibi_scal": 77, "alibi_slop": 77, "alibi_with_scal": 77, "align": [68, 88, 94], "align_corn": 77, "all": [0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 14, 15, 17, 18, 21, 24, 47, 50, 51, 52, 53, 60, 65, 66, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 80, 82, 83, 84, 85, 86, 87, 88, 92, 93, 94], "all_reduce_param": [77, 78], "allbitset": [0, 1], "allgath": [14, 25, 75, 77, 88], "allgeneratedtoken": 0, "alllayersdrafttokenid": 1, "alllayersdrafttokenidspredecessor": 1, "alllayersscor": 1, "alloc": [0, 1, 2, 5, 8, 26, 32, 65, 76, 77, 82, 84, 87, 88, 91, 92, 93, 94], "allocateipcmemori": 1, "allocatespeculativedecodingbuff": 1, "allocnewblock": 0, "allocnewblocksperrequest": 0, "alloctotalblock": 0, "alloctotalblocksperrequest": 0, "allot": 0, "allottedtimem": [0, 88], "allow": [0, 1, 2, 3, 5, 6, 8, 10, 13, 19, 22, 25, 64, 65, 66, 67, 68, 69, 70, 71, 72, 74, 75, 77, 80, 87, 88, 91, 94], "allowed_token_id": 47, "allreduc": [14, 24, 25, 75, 77, 88], "allreducebuff": 1, "allreducefusionkernel": 24, "allreducefusionop": 77, "allreduceparam": [77, 78], "allreducestrategi": 77, "almost": [14, 72, 74, 84], "alon": 4, "along": [5, 10, 16, 60, 77, 88], "alpaca": 9, "alpha": [65, 77, 78, 88], "alphabet": 77, "alreadi": [0, 5, 7, 8, 16, 18, 24, 65, 72, 74, 76, 77, 88, 90, 93], "also": [0, 2, 3, 5, 7, 10, 13, 14, 15, 16, 17, 18, 21, 22, 23, 24, 25, 26, 32, 44, 46, 60, 64, 68, 69, 70, 71, 72, 73, 74, 77, 78, 83, 84, 85, 88, 90, 91, 92, 93], "altair": 88, "alter": [3, 7], "altern": [3, 24, 47, 60, 90, 91], "although": [7, 14, 68, 72, 75], "alwai": [0, 1, 3, 5, 6, 8, 13, 14, 17, 48, 65, 74, 75, 77, 87], "always_share_across_beam": 82, "am": [38, 41, 42, 44, 45, 47, 54, 70, 76, 82], "ambigu": 1, "amd": 88, "amen": [0, 3, 65], "among": [28, 77], "amongst": 77, "amount": [0, 8, 14, 25, 65, 68, 74, 76, 82, 84, 87], "amper": [20, 60, 66, 86, 88], "an": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 19, 21, 23, 24, 25, 26, 32, 38, 40, 41, 42, 43, 44, 45, 47, 54, 60, 62, 64, 65, 66, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 82, 83, 84, 85, 87, 88, 90, 91, 92, 93, 94], "analysi": [7, 24, 59, 84], "analysispatternmanag": 7, "analyt": 20, "analyz": [7, 70], "ani": [0, 1, 2, 3, 7, 10, 15, 17, 18, 26, 47, 60, 64, 65, 66, 68, 69, 74, 75, 76, 77, 79, 82, 87, 90, 91, 92], "announc": [18, 19, 20, 22], "anoth": [0, 1, 5, 7, 9, 17, 20, 24, 26, 74, 77, 87, 92, 94], "answer": 40, "antialia": 77, "antonin": [38, 41, 42, 44, 45], "anybitset": [0, 1], "anyth": [53, 69], "aotman": 88, "apart": 32, "api": [2, 6, 8, 10, 12, 13, 14, 16, 18, 32, 33, 44, 50, 51, 52, 59, 60, 66, 67, 68, 69, 72, 73, 75, 76, 77, 84, 87, 89], "api_kei": [26, 55, 56, 57], "app": [60, 88], "appar": 66, "appear": [0, 5, 6, 46, 62, 65, 77, 87, 88], "append": [47, 54, 67, 77, 94], "append_paged_kv_cach": 92, "appl": 88, "appli": [0, 2, 3, 5, 7, 9, 10, 13, 14, 15, 24, 25, 60, 65, 66, 68, 77, 78, 82, 85, 88, 92], "applic": [8, 10, 20, 23, 24, 26, 29, 30, 31, 62, 64, 66, 67, 83, 87, 88, 94], "apply_batched_logits_processor": [47, 65], "apply_chat_templ": [24, 40], "apply_llama3_sc": 77, "apply_query_key_layer_sc": [78, 79], "apply_residual_connection_post_layernorm": 79, "apply_rotary_pos_emb": 77, "apply_rotary_pos_emb_chatglm": 77, "apply_rotary_pos_emb_cogvlm": 77, "apply_silu": 77, "applybiasropeupdatekvcach": 88, "applyrop": 24, "approach": [2, 4, 7, 8, 10, 24, 64, 68, 76], "appropri": [23, 32, 87], "approxim": [60, 78], "apt": [18, 27, 60, 61, 62], "ar": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 29, 30, 38, 40, 41, 42, 44, 45, 47, 48, 49, 50, 51, 52, 53, 55, 56, 60, 61, 62, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 82, 83, 84, 85, 87, 88, 89, 90, 91, 92, 93, 94], "arang": 77, "arbitrag": 68, "arbitrari": [15, 88], "architectur": [2, 4, 6, 8, 13, 20, 60, 66, 79, 82, 86, 88, 89], "arctic": [86, 88], "area": 54, "aresult": 32, "arg": [7, 17, 26, 49, 65, 78, 79, 82, 88], "arglist": 7, "argmax": 77, "argpars": 49, "argument": [2, 3, 18, 26, 32, 44, 47, 60, 64, 65, 68, 71, 77, 84, 88, 92], "argumentpars": 49, "aris": 60, "arithmet": 14, "armor": 46, "around": [1, 13, 17, 66, 70, 75], "arrai": [0, 1, 65, 77, 82], "arrayview": [0, 1], "arriv": [0, 4], "arrivaltim": 0, "arrow": 77, "art": [18, 24], "articl": [5, 10, 24], "artifici": 66, "artist": 54, "arxiv": [0, 1, 4, 9, 25, 77, 85], "as_dtyp": 77, "as_lay": 7, "as_shap": 77, "ascii": 77, "asciichar": 1, "ask": [46, 53, 87], "aspect": 5, "assembl": [14, 16], "assert": [7, 77, 87, 88, 94], "assert_valid_quant_algo": 79, "assign": [0, 2, 17, 78, 80, 90], "assist": [6, 26, 29, 30, 40, 55, 56, 64, 83], "assistant_model": 6, "associ": [1, 3, 4, 9, 60, 70, 77], "asssembl": 10, "assum": [1, 3, 8, 9, 10, 11, 18, 65, 68, 77, 79, 82], "assumpt": [10, 25], "async": [32, 42, 43, 65, 68, 82], "asynchron": [1, 3, 32, 36, 37, 65], "asyncio": [42, 43], "asyncllmengin": 88, "atom": 1, "attach": [2, 18], "attempt": [0, 2, 69, 70, 72], "attend": 76, "attent": [0, 1, 2, 6, 8, 9, 10, 12, 14, 15, 18, 19, 25, 59, 65, 77, 82, 83, 84, 87, 88, 89, 90, 93], "attention_backend": [90, 92], "attention_head_s": [77, 78], "attention_mask": [77, 78, 79, 82, 92], "attention_mask_param": 79, "attention_mask_typ": 78, "attention_multipli": 79, "attention_output": 87, "attention_output_orig_quant_scal": 77, "attention_output_sf_scal": 77, "attention_packed_mask": [77, 78], "attention_param": [78, 79], "attention_qk_half_accumul": 88, "attentionconfig": 0, "attentionheads": 1, "attentionmask": 92, "attentionmaskparam": 78, "attentionmasktyp": [77, 78], "attentionmetadata": 90, "attentionparam": [78, 79], "attentiontyp": 0, "attn_backend": 92, "attn_bia": 79, "attn_dens": [9, 25], "attn_forward_funcnam": 78, "attn_k": [9, 25], "attn_logit_softcap": 79, "attn_logit_softcapping_scal": 77, "attn_metadata": 90, "attn_processor": 79, "attn_q": [9, 25], "attn_qkv": [9, 25], "attn_v": [9, 25], "attribut": [0, 1, 3, 7, 15, 17, 82], "audio": [82, 88], "audio_engine_dir": 82, "audio_featur": 82, "audio_path": 82, "authent": [64, 70, 83], "authorized_kei": [27, 28], "auto": [0, 1, 2, 3, 5, 6, 11, 14, 38, 46, 65, 68, 75, 77, 79, 80, 81, 88], "auto_deploi": 88, "auto_parallel": [25, 38, 65, 88], "auto_parallel_config": 65, "auto_parallel_world_s": [38, 65], "auto_quantize_bit": 81, "autoawq": 88, "autodeploi": 88, "autogptq": 88, "autom": [40, 88], "automat": [0, 3, 7, 14, 15, 24, 26, 32, 36, 37, 64, 66, 68, 70, 77, 84, 85, 88], "autoparallelconfig": 65, "autopp": 88, "autoq": 88, "autoregress": [0, 10, 92, 93], "autotoken": 32, "autotun": 88, "autotuner_en": 46, "aux": 84, "auxiliari": 10, "avaiable_block": 94, "avail": [0, 1, 3, 7, 8, 14, 19, 21, 26, 32, 38, 41, 42, 44, 45, 47, 60, 66, 68, 74, 75, 76, 82, 83, 84, 85, 88, 89, 92, 93], "averag": [0, 10, 18, 65, 68, 69, 70, 72, 74, 75], "avg": [68, 70, 77], "avg_pool2d": 77, "avgnumdecodedtokensperit": 0, "avgpool2d": 78, "avoid": [1, 2, 17, 24, 60, 64, 82, 84, 88], "awai": [74, 75], "await": [0, 3, 32, 42, 43], "awaitcontextrespons": 0, "awaitgenerationrespons": 0, "awaitrespons": [0, 2, 3], "awar": [2, 5, 19, 87], "awq": [23, 32, 54, 59, 86, 88], "awq_block_s": 81, "ax": 77, "axi": [22, 77], "b": [1, 2, 7, 9, 14, 19, 20, 21, 22, 67, 77, 79, 82, 88], "b200": [69, 88], "b_sf": 77, "back": [0, 2, 8, 10, 41, 44, 62, 69, 88], "backbon": 66, "backend": [0, 2, 3, 10, 14, 16, 18, 26, 33, 40, 46, 47, 50, 51, 52, 59, 65, 67, 68, 69, 83, 88, 91, 93, 94], "backend_token": [0, 3], "backu": [0, 3, 65], "backward": 17, "bad": [0, 3, 65, 88], "bad_token_id": 65, "bad_words_data": 82, "bad_words_list": 82, "badword": 0, "badwordslen": 1, "badwordslist": 1, "badwordsptr": 1, "baichuan": [64, 85, 86, 88], "baichuan2": 86, "baichuanconfig": 79, "baichuanforcausallm": 79, "balanc": [4, 6, 10, 14, 74, 76], "band": 40, "bandwidth": [6, 14, 19, 20, 21, 23, 40], "bangbang": 20, "bantoken": 0, "banword": 0, "bar": 65, "bare": [88, 89], "barissglc": 53, "barnardo": 46, "bart": [86, 88], "base": [0, 1, 2, 3, 8, 9, 10, 12, 15, 16, 17, 18, 19, 20, 23, 24, 25, 42, 43, 49, 60, 65, 66, 68, 74, 76, 77, 78, 79, 80, 81, 82, 84, 86, 88, 89, 90, 91, 93, 94], "base64": 56, "base_model": 9, "base_s": 78, "base_url": [26, 55, 56, 57], "basekvcachemanag": 0, "baselin": [23, 24, 70, 74, 75, 92], "baseline_fp8_engin": 72, "basemodel": 65, "baseresourcemanag": [91, 93], "bash": [14, 26, 28, 29, 30, 31, 33, 34, 35, 50, 51, 52, 67], "basic": [12, 67, 77], "basic_string_view": 0, "batch": [0, 1, 6, 8, 9, 10, 11, 14, 16, 18, 20, 21, 23, 24, 25, 26, 47, 59, 63, 65, 68, 69, 70, 72, 73, 75, 76, 77, 78, 82, 83, 84, 87, 88, 90, 91, 92, 93, 94], "batch_beam_s": [5, 77], "batch_dim": 77, "batch_idx": 82, "batch_input_id": 82, "batch_manag": [0, 1, 93], "batch_schedul": 88, "batch_siz": [5, 7, 11, 13, 19, 22, 77, 78, 81, 82, 84, 92], "batchdon": 1, "batched_logits_processor": [47, 65], "batchedlogitsprocessor": [47, 65], "batchidx": 1, "batchindex": 1, "batching_typ": 65, "batchingtyp": [0, 65], "batchsiz": [0, 1, 6, 20], "batchsizelimit": 0, "batchsizet": 0, "batchslot": 1, "batchslotshostcopi": 1, "batchslotsrequestord": 1, "bc": 77, "beam": [0, 1, 6, 10, 16, 22, 25, 26, 32, 44, 59, 65, 77, 82, 84, 87, 88], "beam_search_diversity_r": [65, 82], "beam_width": [5, 6, 32, 77, 82, 88], "beam_width_arrai": 65, "beamhypothes": 1, "beamsearch": 0, "beamsearchbuff": 1, "beamsearchdiversityr": [0, 1, 6], "beamsiz": 0, "beamtoken": [0, 3], "beamwidth": [0, 1, 2, 3, 6, 65, 88], "beamwidtharrai": [0, 1, 6], "becam": 0, "becaus": [0, 3, 8, 18, 23, 24, 25, 32, 48, 53, 64, 68, 69, 70, 71, 72, 74, 76, 77, 84], "becom": [5, 6, 7, 8, 9, 14, 15, 23, 24, 46, 66], "been": [0, 3, 4, 5, 17, 20, 21, 24, 28, 49, 53, 60, 62, 65, 68, 72, 74, 77, 87, 88], "befor": [0, 1, 2, 3, 5, 7, 8, 9, 13, 14, 15, 24, 50, 51, 52, 59, 60, 62, 66, 67, 71, 72, 74, 76, 77, 79, 82, 84, 87, 88, 90, 91, 92, 93, 94], "beforehand": 70, "begin": [10, 64, 66, 71, 88, 90], "behav": [0, 84], "behavior": [2, 5, 69, 74, 77, 82, 84, 88], "behaviour": [0, 77], "behind": 20, "being": [0, 5, 8, 14, 17, 53, 65, 74, 87, 88, 92], "believ": [46, 68], "belong": 74, "below": [0, 5, 6, 7, 9, 18, 21, 22, 23, 27, 28, 68, 69, 72, 74, 75, 87], "bench": [18, 36, 37, 53, 68, 69, 73, 88], "benchmark": [24, 51, 59, 60, 67, 72, 73, 75, 83, 88], "benchmark_2nod": 26, "benefici": [68, 74, 75], "benefit": [7, 8, 21, 23, 25, 66, 74, 88], "bert": [25, 77, 85, 86, 88], "bert_attent": 77, "bert_attention_plugin": 25, "bert_context_fmha_fp32_acc": 25, "bertattent": 78, "bertattentionplugin": 77, "bertbas": 79, "bertforquestionansw": 79, "bertforsequenceclassif": [79, 86], "bertmodel": 79, "besid": 91, "best": [5, 14, 24, 48, 65, 67, 68, 71, 73, 74, 83, 88], "best_of": [65, 88], "best_path": 82, "best_path_len": 82, "best_path_length": 82, "best_perf_practice_on_deepseek": [24, 88], "bestpathindic": 1, "bestpathlength": 1, "beta": [26, 77], "beta_fast": 77, "beta_slow": 77, "better": [0, 2, 5, 6, 8, 15, 17, 22, 24, 25, 50, 51, 52, 65, 69, 71, 72, 75, 76, 88], "between": [0, 2, 5, 6, 8, 10, 14, 15, 17, 24, 30, 56, 62, 65, 67, 69, 71, 75, 76, 77, 78, 84, 87, 88, 90], "beyond": [1, 20, 72], "bf16": [5, 15, 17, 24, 59, 72, 75, 86, 88], "bfloat16": [5, 14, 25, 68, 70, 80, 85, 86, 88], "bhuvanesh09": 88, "bi": 5, "bia": [0, 3, 13, 14, 65, 77, 78, 79, 88], "bias": [13, 77], "bidirect": [77, 78], "bidirectionalglm": 77, "bigger": 8, "biggest": 8, "billion": 18, "bin": [13, 14, 15, 18, 26, 29, 30, 31, 33, 34, 35, 50, 51, 52, 67, 87, 88], "binari": [10, 14, 67, 77], "bind": [47, 59, 65, 76, 82, 84, 88, 91, 93, 94], "bindcapacityschedul": 94, "bit": [0, 1, 5, 20, 53, 77, 85], "bitmask": 88, "bl": [10, 79], "black": 7, "blackwel": [2, 18, 54, 59, 62, 71, 72, 86, 88], "blip": [85, 88], "blip2": [85, 86, 88], "blob": 24, "block": [0, 1, 2, 5, 6, 8, 14, 25, 32, 46, 47, 64, 65, 74, 77, 82, 84, 88, 93], "block_controlnet_hidden_st": 79, "block_hash": 46, "block_num": 77, "block_siz": [77, 78, 82], "block_sparse_block_s": 77, "block_sparse_homo_head_pattern": 77, "block_sparse_num_local_block": 77, "block_sparse_param": 78, "block_sparse_vertical_strid": 77, "blockhash": 0, "blockidx": 1, "blockptr": 1, "blocksiz": 0, "blockspars": 77, "blocksparseattnparam": 78, "blog": [18, 19, 22, 23, 24, 88], "bloodeagle40234": 88, "bloom": [6, 15, 85, 86, 88], "bloom_dict": 15, "bloomforcausallm": 79, "bloommodel": 79, "bm": 1, "bmm": 14, "board": 75, "bodi": 14, "book": 53, "bool": [0, 1, 7, 11, 13, 65, 77, 78, 79, 80, 82, 92], "boolean": [1, 3, 9, 77, 79, 80], "boost": [18, 24, 72, 74, 75], "born": [12, 14, 87], "borrow": [32, 44, 68], "bos_token_id": 82, "both": [0, 2, 4, 5, 7, 9, 10, 14, 15, 18, 20, 23, 24, 25, 36, 49, 65, 68, 69, 71, 74, 76, 77, 78, 84, 85, 88, 91, 92], "bottleneck": [4, 18, 23, 71, 74], "bottom": 28, "bound": [0, 6, 12, 14, 21, 24, 65, 68, 77, 82, 84], "boundari": [6, 14, 65, 77, 79, 81, 84], "box": [7, 18], "bpru": 88, "brahma": 68, "branch": [10, 19, 22, 65], "breadth": 10, "break": [10, 24, 64, 68, 75, 88, 94], "breakdown": [67, 68, 69, 70], "breviti": 18, "brief": [79, 82, 92], "briefli": [30, 56], "brife": 0, "bring": [23, 24, 90], "broadcast": [3, 24, 77], "broadcast_help": 77, "broader": [5, 88], "broken": [66, 74, 88], "bsz": 78, "bu": 60, "budget": [11, 74], "buffer": [0, 1, 2, 3, 8, 25, 26, 59, 65, 77, 88, 93], "buffer_0": 1, "buffer_1": 1, "buffer_2": 1, "buffer_alloc": 82, "buffercast": 1, "buffercastornul": 1, "bufferdatatyp": 1, "buffermanag": 84, "buffermanagertest": 1, "bufferptr": 1, "bufferrang": 1, "buffers": 1, "bufferview": 0, "bug": 88, "build": [2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 14, 16, 39, 44, 46, 48, 49, 53, 59, 64, 65, 66, 67, 71, 72, 73, 74, 76, 79, 80, 83, 84, 87, 88], "build_cach": 65, "build_config": [17, 25, 32, 39, 44, 48, 49, 53, 65, 72, 74, 75, 79], "build_dir": 60, "build_engin": 14, "build_flags_multiple_profil": 75, "build_serialized_network": 14, "build_wheel": [18, 60, 67], "buildcacheconfig": 65, "buildconfig": [11, 17, 32, 39, 44, 48, 49, 53, 65, 72, 74, 75, 88], "builder": [11, 14, 17, 65, 88], "builder_force_num_profil": 88, "builder_opt": 88, "built": [3, 6, 8, 14, 17, 25, 54, 60, 62, 64, 65, 68, 69, 70, 75, 76, 77, 83, 84, 87, 88], "bump": 1, "bumptaskinprogress": 1, "burden": 71, "busi": 0, "button": 88, "buvnswrn": 88, "bw": 88, "byt5": [86, 88], "byte": [0, 1, 65, 82], "bytestostr": 1, "c": [0, 1, 2, 5, 7, 10, 14, 16, 18, 26, 27, 28, 32, 50, 51, 52, 59, 65, 66, 67, 74, 77, 79, 83, 88, 91, 93, 94], "cach": [0, 1, 2, 3, 6, 9, 14, 17, 23, 24, 25, 26, 32, 36, 37, 39, 49, 59, 63, 65, 66, 68, 69, 70, 74, 77, 82, 83, 85, 88, 89, 90, 91, 92, 94], "cache_indir": 82, "cache_indir_t": 77, "cache_indirect": [5, 77, 78, 82, 87], "cache_root": 65, "cache_transceiver_config": 65, "cachehitr": 0, "cacheindirect": 1, "cachelevel": 0, "cachelevelupd": 0, "caches": 0, "cachest": 0, "cachetransceiv": 0, "cachetransceiverconfig": [0, 65], "cachetyp": 93, "cachevalu": 1, "calcul": [0, 19, 20, 22, 65, 68, 76, 77, 82, 84, 88], "calculate_speculative_resourc": 65, "calculatespeculativeresourc": 0, "calculatespeculativeresourcetupl": 0, "calib_batch": [54, 65, 72, 79], "calib_batch_s": [65, 72, 79], "calib_config": [54, 65, 72], "calib_dataset": [54, 65, 79, 81], "calib_max_seq_length": [54, 65, 72, 79, 81], "calib_s": [68, 81], "calibconfig": [54, 65, 72], "calibr": [15, 23, 25, 54, 65, 72, 88], "call": [0, 1, 3, 4, 5, 6, 7, 14, 15, 17, 32, 47, 65, 67, 70, 72, 77, 79, 81, 82, 83, 84, 88, 90, 91, 92, 93], "callabl": [15, 47, 65, 79], "callback": [3, 47, 65], "can": [0, 1, 2, 3, 4, 5, 6, 7, 10, 11, 14, 15, 16, 17, 18, 19, 20, 22, 23, 24, 25, 26, 27, 28, 32, 36, 39, 41, 44, 47, 48, 49, 50, 51, 52, 53, 54, 59, 60, 62, 64, 65, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 79, 80, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94], "canaccessp": 1, "cancel": [0, 3, 65, 68, 88], "cancelrequest": [0, 3], "candid": [0, 6, 10, 14, 24], "canenqueu": 0, "canenqueuerequest": 0, "cannon": 46, "cannot": [1, 6, 14, 15, 24, 64, 65, 74, 75, 76, 77, 84, 87, 88, 94], "cap": 70, "capabl": [19, 24, 40, 60, 66, 67, 72], "capac": [0, 1, 19, 21, 23, 65, 94], "capacitor_schedul": 94, "capacity_scheduler_polici": [65, 76], "capacityschedul": [91, 93, 94], "capacityschedulerpolici": [0, 65, 76, 88], "capit": [36, 38, 39, 41, 42, 43, 44, 45, 49, 54, 61, 62, 70, 76, 83, 89], "caption": 78, "captur": [65, 92], "card": [48, 53], "carefulli": 18, "case": [0, 1, 2, 5, 6, 8, 9, 10, 18, 20, 23, 24, 25, 32, 68, 69, 70, 72, 73, 75, 77, 85, 88], "cast": 77, "cast_to_dtyp": 77, "castsiz": 1, "cat": [18, 26, 51], "categor": [10, 77], "categori": 80, "categorical_sampl": 77, "caus": [2, 3, 15, 17, 25, 65, 75, 87, 88], "causal": [77, 78, 92], "cautiou": 17, "caveat": 72, "cd": [12, 13, 18, 60, 68, 83, 87, 89], "ceil": [1, 79], "ceil_mod": [77, 78], "ceildiv": 1, "center": [20, 21], "central": 80, "certain": [2, 7, 13, 62, 66, 77], "cg": 79, "challeng": [24, 66], "chanc": [8, 25, 76], "chang": [2, 5, 6, 8, 9, 15, 17, 19, 21, 22, 60, 64, 65, 66, 68, 75, 77, 79, 82, 84, 87, 89, 93], "channel": [25, 77, 85, 88], "char": [0, 1], "charg": [6, 14, 92], "chart": 20, "chat": [10, 21, 31, 34, 36, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48, 53, 54, 57, 58, 61, 62, 64, 83, 88, 89], "chatbot": 53, "chatcmpl": 83, "chatglm": [64, 77, 85, 86, 88], "chatglm2": [64, 86, 88], "chatglm3": [64, 79, 86, 88], "chatglm_vers": 79, "chatglmconfig": 79, "chatglmforcausallm": 79, "chatglmgenerationsess": 82, "chatglmmodel": 79, "check": [2, 3, 36, 61, 62, 65, 69, 71, 72, 74, 75, 77, 82, 83, 84, 87, 88, 90], "check_accuraci": 13, "check_config": 79, "check_gpt_mem_usag": 84, "checkbeamsearchdiversityr": 0, "checkbeamwidth": 0, "checkbeamwidtharrai": 0, "checkearlystop": 0, "checklengthpenalti": 0, "checkminp": 0, "checkmintoken": 0, "checknorepeatngrams": 0, "checknumreturnsequ": 0, "checkpoint": [12, 15, 16, 17, 18, 24, 25, 26, 41, 49, 59, 64, 65, 68, 70, 72, 81, 82, 83, 85, 87, 88, 90], "checkpoint_dir": [9, 11, 12, 13, 14, 17, 25, 68, 83, 87], "checkposteriorvalu": 0, "checkrepetitionpenalti": 0, "checktemperatur": 0, "checktopk": 0, "checktopp": 0, "checktoppdecai": 0, "checktoppmin": 0, "checktoppresetid": 0, "chef": 87, "chmod": 27, "choic": [0, 10, 23, 25, 49, 68, 71, 77, 82, 83, 92], "choos": [14, 17, 24, 72, 77, 88], "chosen": [84, 94], "chrome": 67, "chrono": 0, "chunk": [0, 25, 59, 63, 65, 75, 77, 82, 84, 88], "chunk_dim": 78, "chunk_length": 88, "chunk_scan": 77, "chunk_siz": [77, 79], "chunkedcontextnexttoken": 1, "chunkedcontextnexttokenshost": 1, "ci": 1, "circular": 5, "citi": [54, 83], "ckpt": [49, 68, 83], "ckpt_dir": [14, 17, 79], "ckpt_llama_3": 14, "cl": [12, 17], "claim": [1, 15], "claimpag": 1, "claimpageswithevict": 1, "clamp": [65, 88], "clamp_val": 65, "class": [0, 1, 2, 5, 6, 7, 11, 12, 14, 15, 17, 23, 25, 32, 39, 41, 44, 47, 48, 49, 60, 64, 65, 71, 72, 75, 77, 78, 79, 80, 81, 82, 87, 88, 90, 91, 92, 94], "class_dropout_prob": 78, "class_label": 78, "classic": [14, 59], "classifi": [78, 79], "classmethod": [12, 17, 65, 78, 79, 82], "classvar": 65, "clean": [18, 60, 67, 87], "clear": [62, 74, 82], "clearli": 76, "cli": [13, 18, 32, 59, 68, 71, 72, 74, 75, 83], "click": [27, 28], "client": [0, 3, 26, 58, 69], "client_id": 47, "clientid": 0, "clip": 77, "clip_before_cast": 77, "clip_qkv": [78, 79], "clip_vision_model": 79, "clipvisiontransform": 79, "clock": 24, "clone": [9, 18, 60, 64, 70, 83, 87, 89], "clone_input": 7, "close": [5, 17, 18, 25, 75, 84], "closur": 77, "cloud": [20, 27, 28], "cls_token": 78, "cluster": [6, 14, 24, 25, 26, 62, 65, 88], "cluster_info": 88, "cluster_kei": [25, 88], "cluster_s": 26, "cmake": [60, 88], "cnn_dailymail": [54, 65, 79], "co": [0, 9, 18, 30, 56, 64, 77, 78, 83, 87], "coalesc": 47, "coast": 83, "code": [2, 5, 7, 10, 14, 17, 23, 24, 26, 32, 50, 51, 52, 59, 64, 65, 66, 67, 68, 77, 85, 86, 87, 88, 90, 93, 94], "codebas": 90, "codellama": 88, "codepath": 88, "codeqwen": 88, "coderham": 88, "cogvlm": [86, 88], "cogvlmattent": 78, "cogvlmconfig": 79, "cogvlmforcausallm": 79, "coher": [6, 88], "cohereconfig": 79, "cohereforcausallm": 79, "collabor": [6, 24, 54, 77], "collect": [1, 7, 10, 14, 24, 65, 69, 77, 90], "collect_and_bia": 78, "color": [53, 74], "column": [9, 77, 85], "columnlinear": [9, 12, 78], "com": [17, 18, 24, 60, 77, 83, 87, 88, 89], "combin": [0, 7, 10, 21, 24, 25, 49, 50, 51, 52, 68, 69, 72, 74, 78, 88, 92, 94], "combinedtimesteplabelembed": 78, "combinedtimesteptextprojembed": 78, "come": [6, 9, 20, 70, 71, 74, 76, 84, 87], "comm": 65, "comma": [77, 82], "command": [8, 9, 12, 13, 14, 17, 18, 26, 27, 28, 50, 51, 52, 60, 64, 67, 68, 70, 75, 80, 83, 84, 87, 88, 89], "commandr": 88, "comment": 88, "commmod": 0, "common": [0, 5, 8, 10, 18, 36, 46, 64, 77, 84, 93], "common_prefix": 46, "commonli": [7, 24, 26, 88], "commstat": 0, "commtyp": 0, "commun": [0, 2, 6, 14, 25, 54, 64, 66, 72, 77, 86, 88], "communicationmod": [0, 2], "communicationtyp": 0, "compani": 48, "compar": [1, 2, 15, 20, 21, 23, 72, 74, 75, 76, 77, 92], "comparison": [6, 20, 24, 68], "compat": [10, 17, 26, 60, 75, 78, 83, 86, 88, 90], "compbin": 9, "compil": [6, 16, 59, 62, 66, 67, 68, 77, 87], "complet": [0, 1, 2, 3, 6, 8, 10, 29, 30, 32, 55, 56, 58, 60, 64, 65, 66, 68, 69, 70, 74, 75, 83, 88, 93, 94], "completion_token": 83, "completionoutput": [32, 48, 65], "complex": [7, 10, 14, 24], "compli": 26, "complic": 90, "compon": [2, 3, 5, 14, 16, 23, 24, 59, 85, 91], "compos": [0, 6, 68], "comprehens": [18, 26, 66], "compress": 19, "compris": 23, "comput": [0, 1, 4, 5, 6, 8, 10, 14, 19, 20, 21, 23, 24, 25, 38, 41, 42, 44, 45, 47, 67, 68, 71, 72, 76, 77, 84, 87, 88, 90, 91, 92, 93], "compute_relative_bia": 78, "computecontextlogit": 1, "computegenerationlogit": 1, "computenumpackedmask": 1, "concat": [12, 24, 77], "concat_kvcach": 24, "concaten": [5, 9, 15, 24, 77, 90], "conced": 46, "concept": [14, 68, 73, 88, 93], "conceptu": 1, "concern": [14, 84], "conclus": 73, "concret": 90, "concur": 46, "concurr": [1, 2, 10, 18, 20, 24, 68, 88], "cond_proj_dim": 78, "conda": 88, "condit": [0, 1, 3, 6, 7, 10, 68, 77, 78, 88], "condition": 77, "conditioning_embed": 78, "conditioning_embedding_dim": 78, "conduct": [5, 68], "confess": 46, "config": [0, 1, 5, 8, 9, 11, 12, 15, 17, 18, 19, 26, 33, 46, 65, 68, 74, 78, 79, 80, 82, 87, 88, 90, 93], "config_class": 79, "config_dir": 79, "config_fil": [26, 65, 79], "configdict": 65, "configur": [0, 1, 2, 4, 5, 10, 15, 16, 18, 21, 25, 26, 39, 40, 44, 48, 49, 53, 60, 62, 65, 68, 69, 70, 73, 74, 76, 79, 82, 84, 87, 88, 92], "configuration_llama": 90, "configuration_mymodel": 90, "configuration_util": 90, "confirm": [38, 41, 42, 44, 45], "conform": 65, "conjunct": 74, "connect": [0, 14, 70, 71, 73], "connectionmanag": 0, "consecut": 6, "consequ": [2, 23, 71, 75], "conserv": [0, 76], "consid": [0, 1, 9, 10, 18, 23, 53, 54, 65, 69, 74, 77, 90, 94], "consider": [17, 23, 32], "consist": [7, 17, 20, 24, 66, 68, 70, 77, 85, 87, 92], "consol": 27, "consolid": 10, "const": [0, 1, 3], "const_iter": 1, "constant": [1, 5, 77, 84], "constant_to_tensor_": 77, "constantli": [38, 41, 42, 44, 45], "constants_to_tensors_": 77, "constantthreshold": 1, "constexpr": [0, 1], "constpointercast": 1, "constrain": [6, 23], "constraint": [0, 5, 6, 23, 62, 77], "construct": [0, 1, 3, 10, 14, 68, 77, 88, 92], "constructor": [0, 11, 53, 64, 83, 92], "consult": [10, 60, 67], "consum": [0, 7, 65, 77], "consumpt": [5, 20, 25], "contact": 77, "contain": [0, 1, 2, 3, 5, 6, 7, 9, 13, 14, 15, 16, 17, 24, 25, 26, 28, 50, 51, 52, 61, 62, 65, 66, 68, 69, 77, 79, 82, 83, 85, 86, 88, 89, 91, 92], "container_imag": [50, 51, 52], "container_img": 26, "content": [1, 9, 17, 26, 27, 29, 30, 31, 40, 55, 56, 59, 77, 83, 84, 88], "context": [0, 2, 4, 8, 23, 25, 59, 63, 65, 68, 73, 77, 82, 84, 87, 88, 92, 93, 94], "context_chunking_polici": [65, 76], "context_fmha": [9, 25], "context_fmha_fp32_acc": 88, "context_fmha_typ": [5, 84], "context_init": 94, "context_len": [82, 92], "context_length": [77, 78, 82, 87], "context_logit": [65, 82], "context_mem_s": 82, "context_onli": 65, "context_parallel_s": 65, "context_phas": 5, "context_pre_onli": 78, "context_request": 94, "contextchunkingpolici": [0, 65, 76, 88], "contextexecutor": 2, "contextfmha": 1, "contextidx": 0, "contextlogit": 0, "contextmanag": 64, "contextparallel": 1, "contextphaseparam": [0, 2, 65], "contextpositionid": 1, "contextprefillposit": 0, "contextrequest": 1, "contextrequestid": 2, "contextrespons": 2, "contigu": [2, 71, 77, 88], "continu": [1, 3, 5, 10, 21, 23, 25, 65, 66, 72, 74, 82, 94], "contract": 68, "contrast": [6, 10, 92], "contrib": 19, "contribut": [17, 68, 77, 88], "contributor": [24, 84], "control": [0, 2, 5, 6, 7, 32, 36, 37, 65, 67, 68, 70, 76, 77, 78, 82, 85, 88], "conv": 77, "conv1d": [25, 77, 78], "conv2d": [77, 78], "conv3d": [77, 78], "conv_bia": 77, "conv_kernel": 82, "conv_stat": 79, "conv_state_or_ptr": 77, "conv_transpose2d": 77, "conv_weight": 77, "conveni": [1, 12, 17, 60], "convent": [17, 77], "convers": [1, 15, 22, 23, 53, 59, 83, 88], "convert": [0, 1, 9, 11, 12, 13, 14, 15, 17, 66, 68, 70, 72, 83, 87, 88, 92], "convert_and_load_weights_into_trtllm_llama": 17, "convert_checkpoint": [9, 11, 12, 13, 14, 17, 70, 71, 83, 87, 88], "convert_coneckpoint": 4, "convert_hf_mpt_legaci": 88, "convert_util": 88, "convert_weights_from_custom_training_checkpoint": 17, "convkernel": 1, "convolut": [0, 82], "convtranspose2d": 78, "coordin": [10, 59, 77], "copi": [0, 1, 2, 8, 10, 25, 28, 65, 72, 77, 84, 88, 92], "copy_on_partial_reus": 65, "copyfrom": 1, "copyonpartialreus": 0, "copytask": 1, "copytaskmappag": 1, "copyto": 0, "copytocpu": 0, "copytogpu": 0, "copytomanag": 0, "copytopag": 1, "copytopin": 0, "copytopooledpin": 0, "core": [6, 7, 9, 11, 14, 17, 19, 20, 22, 60, 64, 68, 71, 83, 87, 88, 91], "coroutin": [42, 43, 65], "correct": [2, 3, 5, 9, 10, 88], "correctli": [8, 77, 88, 90], "correspond": [0, 1, 2, 4, 5, 7, 9, 10, 15, 17, 26, 65, 67, 75, 77, 78, 82, 85, 87, 88, 90], "cost": [8, 14, 24, 68, 71, 84, 88], "costli": 24, "could": [0, 2, 7, 8, 13, 41, 42, 43, 44, 45, 54, 65, 70, 84, 87, 88], "couldn": 74, "count": [0, 1, 6, 26, 34, 35, 64, 68, 79, 83], "count_include_pad": [77, 78], "countlocallay": 1, "countlowerranklay": 1, "cours": 10, "court": [38, 41, 42, 44, 45], "cover": [18, 72, 73, 75], "cp312": 60, "cp_config": 65, "cp_group": [77, 78], "cp_rank": [77, 78], "cp_size": [77, 78, 81, 88], "cp_split_plugin": 77, "cpp": [2, 3, 5, 6, 14, 18, 26, 51, 59, 60, 67, 68, 69, 70, 87, 88], "cpp_e2e": 82, "cpp_extens": 62, "cpp_llm_onli": 82, "cpp_onli": 60, "cpu": [0, 1, 8, 9, 11, 14, 24, 25, 26, 47, 62, 65, 77, 84, 87, 88, 92], "cpumemusag": [0, 65], "crash": 88, "creat": [1, 2, 3, 7, 8, 10, 11, 12, 14, 16, 17, 24, 26, 27, 32, 38, 41, 42, 43, 44, 45, 46, 47, 54, 55, 56, 57, 64, 65, 66, 68, 69, 70, 74, 75, 77, 78, 79, 82, 83, 84, 88, 90, 91, 92, 94], "create_allreduce_plugin": 77, "create_attention_const_param": 78, "create_builder_config": 11, "create_cuda_graph_metadata": 92, "create_execution_context": 82, "create_fake_weight": 77, "create_network": 14, "create_pytorch_model_based_executor": [93, 94], "create_runtime_default": 79, "create_sinusoidal_posit": 77, "create_sinusoidal_positions_for_attention_plugin": 77, "create_sinusoidal_positions_for_cogvlm_attention_plugin": 77, "create_sinusoidal_positions_long_rop": 77, "create_sinusoidal_positions_yarn": 77, "createloramodul": 1, "creation": [1, 65, 77, 84], "creativ": 6, "criteria": 82, "critic": [24, 68, 87], "crop": 78, "cropped_pos_emb": 78, "cross": [0, 9, 24, 65, 77, 82, 88], "cross_attent": [78, 82], "cross_attention_dim": 78, "cross_attention_mask": [78, 82], "cross_attention_mask_for_context": 82, "cross_attention_mask_for_gen": 82, "cross_attention_norm": 78, "cross_attention_norm_num_group": 78, "cross_attention_packed_mask": 78, "cross_attn_dens": [9, 25], "cross_attn_k": [9, 25], "cross_attn_q": [9, 25], "cross_attn_qkv": [9, 25], "cross_attn_v": [9, 25], "cross_kv": 77, "cross_kv_cache_block_offset": [78, 82], "cross_kv_cache_fract": [65, 82], "cross_kv_cache_gen": [78, 79], "cross_kv_length": 77, "cross_kv_reus": [78, 79], "crossattentionmask": 0, "crosskvcachefract": [0, 88], "crosskvcachestat": 0, "crucial": [10, 14, 23, 91], "ctor": 77, "ctx": 0, "ctx_request_id": 65, "ctxenginepath": 0, "ctxexecutorconfig": 0, "cu": [14, 24], "cu12": 88, "cu128": [61, 62], "cuassert": 87, "cublaslt": [25, 75], "cuda": [0, 1, 2, 5, 14, 18, 47, 54, 60, 61, 62, 65, 67, 68, 79, 82, 84, 87, 88, 92, 93], "cuda_arch": 60, "cuda_architectur": [18, 60], "cuda_graph_batch_s": [18, 69], "cuda_graph_cache_s": 65, "cuda_graph_inst": 87, "cuda_graph_mod": [65, 82, 87], "cuda_graph_padding_en": [18, 51, 69], "cuda_hom": 62, "cuda_launch_block": 87, "cuda_stream": 87, "cuda_stream_guard": 82, "cuda_stream_sync": 77, "cudadevicegetstreampriorityrang": 1, "cudaevent_t": 1, "cudaeventdisabletim": 1, "cudagraph": 88, "cudagraphcaches": 0, "cudagraphlaunch": 87, "cudagraphmod": 0, "cudamalloc": [1, 2], "cudamallocasync": [1, 2], "cudamemcpyasync": 47, "cudamempool": 1, "cudamempoolptr": 1, "cudaprofilerapi": 67, "cudart": 87, "cudastream": 0, "cudastream_t": 1, "cudastreamcreatewithflag": 1, "cudastreamnonblock": 1, "cudastreamptr": [0, 1], "cudeviceptr": 1, "cudnn": 88, "cumemgenericallocationhandl": 1, "cumlogprob": [0, 1], "cumlogprobscba": 1, "cumsum": [77, 88], "cumsumgenerationlength": 1, "cumsumlastdim": 77, "cumsumlength": 1, "cumul": [0, 1, 65, 77], "cumulative_logprob": [32, 48, 65], "curand": 88, "curl": [26, 58, 83], "currenc": 68, "current": [0, 1, 2, 3, 5, 9, 10, 18, 23, 24, 25, 32, 40, 53, 60, 68, 72, 74, 75, 76, 77, 82, 84, 86, 88, 89, 91, 92, 93, 94], "current_stream": 87, "currentexpandindic": 1, "curv": 22, "custom": [6, 14, 17, 19, 24, 25, 36, 37, 39, 47, 48, 49, 60, 66, 72, 75, 77, 82, 88, 91, 92], "custom_all_reduc": 88, "custom_mask": 77, "customallreduc": 88, "customized_key_dict": 15, "customized_preprocess": 15, "customizedmodulea": 15, "customizedmoduleb": 15, "cutlass": 88, "cxx11": 60, "cyclic": [59, 77, 82], "d": [1, 9, 26, 27, 29, 30, 31, 50, 51, 52, 53, 68, 77, 78, 83, 87, 88], "d0": 24, "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b": 68, "dangl": 7, "data": [0, 1, 2, 5, 6, 14, 15, 19, 20, 21, 22, 23, 24, 25, 46, 56, 65, 68, 69, 70, 77, 79, 86, 87, 88, 90], "data_path": 51, "data_typ": [11, 13], "datacontext": 0, "dataset": [24, 30, 51, 54, 56, 65, 67, 72, 88], "dataset_fil": 69, "dataset_path": 68, "datatyp": [0, 1, 6, 14, 77, 82, 85, 87], "datatypetrait": 1, "date": 17, "datetim": 65, "dbrx": [85, 86, 88], "dbrxconfig": 79, "dbrxforcausallm": 79, "dconv": 77, "de": 1, "deactiv": 32, "dead": 88, "deal": [5, 7, 87], "dealloc": [1, 94], "death": [38, 41, 42, 44, 45], "debug": [0, 25, 26, 59, 60, 82, 84, 88], "debug_buff": 87, "debug_mod": [82, 87], "debug_tensors_to_sav": 82, "debugconfig": 0, "debuginputtensor": 0, "debugoutputtensor": 0, "debugtensor": 0, "debugtensornam": 0, "debugtensorsmaxiter": 0, "debugtensorsperiter": 0, "dec": [25, 82, 88], "decai": [0, 6, 65], "decid": [5, 13, 59, 68, 73, 74, 85, 91, 94], "decilmforcausallm": 86, "decis": [53, 77], "declar": [1, 6, 7, 17, 91, 93], "decltyp": [0, 1], "decod": [0, 1, 2, 5, 6, 12, 17, 24, 26, 36, 37, 59, 65, 68, 77, 82, 86, 88, 90, 93], "decode_batch": 82, "decode_duration_m": 65, "decode_regular": 82, "decode_retention_prior": 65, "decode_stream": 82, "decode_words_list": 82, "decode_wrapp": 92, "decodedurationm": 0, "decoder_batch": 1, "decoder_input_id": [79, 82], "decoder_language_adapter_rout": 82, "decoder_lay": 90, "decoder_start_token_id": 25, "decoderbuff": 1, "decoderenginebuff": 0, "decoderetentionprior": 0, "decoderjsonconfigstr": 0, "decoderlay": 90, "decoderlayerlist": 12, "decoderlookaheadbuff": 1, "decodermaskedmultiheadattent": 5, "decodermodel": [0, 79, 90], "decodermodelforcausallm": [12, 17, 79, 90], "decodermodelpath": 0, "decoderst": 88, "decoderxqarunn": 5, "decoding_typ": [18, 65], "decodingbaseconfig": 65, "decodingconfig": [0, 1], "decodinginputptr": 1, "decodingit": 0, "decodinglayerworkspac": 1, "decodingmod": [0, 1, 88], "decodingoutputptr": 1, "decompos": 5, "decor": 90, "decoupl": [24, 84], "decreas": [19, 20, 72], "dedic": [24, 87], "deduc": [25, 26, 88], "deep": [14, 20, 21, 67, 77, 88], "deepgemm": 18, "deeplearn": [77, 87], "deepseek": [26, 58, 67, 69, 86, 88], "deepseek_v1": 88, "deepseek_v2": 88, "deepseek_v3": [24, 88], "deepseekforcausallm": 79, "deepseekv1config": 79, "deepseekv2": 77, "deepseekv2attent": 78, "deepseekv2config": 79, "deepseekv2forcausallm": 79, "deepseekv3forcausallm": 86, "deepspe": 13, "def": [7, 12, 14, 15, 17, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 61, 62, 70, 72, 75, 76, 83, 87, 89, 90, 94], "default": [0, 1, 2, 3, 4, 5, 6, 8, 13, 15, 17, 25, 26, 27, 32, 49, 54, 59, 60, 65, 67, 69, 72, 73, 74, 75, 76, 77, 79, 82, 83, 84, 85, 87, 88, 90, 92], "default_net": 77, "default_plugin_config": 79, "default_trtnet": 14, "defaultvalu": 1, "defer": 77, "defin": [0, 1, 3, 5, 7, 10, 13, 14, 15, 16, 17, 18, 21, 25, 66, 68, 75, 77, 78, 85, 88, 90, 92], "definit": [3, 5, 16, 17, 24, 59, 66, 77, 87], "deftruth": 88, "degrad": [0, 2, 25, 72], "degre": [38, 41, 42, 44, 45, 47, 69, 72, 75], "delai": [69, 88], "deleg": [77, 92], "delet": [0, 1, 80, 87], "deliv": [18, 19, 22, 24, 69], "delta": [0, 24, 77, 78], "delta_bia": 77, "delta_softplu": 77, "demand": 24, "demo": [24, 30, 56], "demonstr": [3, 15, 20, 24, 64, 70, 72, 74, 75], "denmark": 46, "denois": 78, "denot": 10, "dens": [4, 5, 9, 13, 15, 77], "dense_4h_to_h": 15, "dense_bia": 78, "dense_h_to_4h": 15, "densiti": 23, "dep": 60, "departur": 46, "depend": [0, 2, 3, 5, 6, 7, 10, 13, 21, 26, 62, 65, 69, 70, 72, 75, 77, 84, 87, 88, 93], "deploi": [10, 13, 26, 59, 62, 66], "deplot": [86, 88], "deploy": [23, 24, 66, 68, 72, 83, 88], "deprec": [25, 66, 68, 88], "deprecationwarn": 68, "depriv": 7, "depth": 10, "dequ": [0, 1], "dequant": [5, 59, 77], "deriv": [14, 15, 77, 84, 91], "descendli": 6, "describ": [5, 6, 8, 9, 10, 12, 14, 15, 16, 18, 22, 28, 30, 56, 60, 64, 68, 69, 75, 77, 85, 87, 92], "descript": [0, 1, 6, 9, 26, 49, 59, 68, 69, 75, 77, 92], "deseri": 17, "deserializeadditionalmodeloutput": 0, "deserializeadditionaloutput": 0, "deserializebool": 0, "deserializecachest": 0, "deserializecachetransceiverconfig": 0, "deserializecommst": 0, "deserializecontextphaseparam": 0, "deserializedatatransceiverst": 0, "deserializedebugconfig": 0, "deserializedecodingconfig": 0, "deserializedecodingmod": 0, "deserializedisservingrequeststat": 0, "deserializedynamicbatchconfig": 0, "deserializeeagleconfig": 0, "deserializeexecutorconfig": 0, "deserializeextendedruntimeperfknobconfig": 0, "deserializeexternaldrafttokensconfig": 0, "deserializeguideddecodingconfig": 0, "deserializeguideddecodingparam": 0, "deserializeinflightbatchingstat": 0, "deserializeiterationstat": 0, "deserializeiterationstatsvec": 0, "deserializekvcacheconfig": 0, "deserializekvcacheretentionconfig": 0, "deserializekvcachestat": 0, "deserializelookaheaddecodingconfig": 0, "deserializeloraconfig": 0, "deserializemodeltyp": 0, "deserializemropeconfig": 0, "deserializeorchestratorconfig": 0, "deserializeoutputconfig": 0, "deserializeparallelconfig": 0, "deserializepeftcacheconfig": 0, "deserializeprompttuningconfig": 0, "deserializerequest": 0, "deserializerequestperfmetr": 0, "deserializerequeststag": 0, "deserializerequeststat": 0, "deserializerequeststatsperiter": 0, "deserializerequeststatsperiterationvec": 0, "deserializerespons": 0, "deserializeresult": 0, "deserializesamplingconfig": 0, "deserializeschedulerconfig": 0, "deserializesocketst": 0, "deserializespecdecfastlogitsinfo": 0, "deserializespeculativedecodingconfig": 0, "deserializestaticbatchingstat": 0, "deserializestr": 0, "deserializetensor": 0, "deserializetimepoint": 0, "deserializetokenrangeretentionconfig": 0, "design": [1, 10, 14, 15, 17, 18, 23, 24, 64, 70, 83, 91, 92, 93], "desir": [3, 69, 77, 83, 92], "destin": [50, 51, 52], "destroi": [1, 84], "destroyipcmemori": 1, "destructor": 1, "detail": [0, 3, 5, 10, 12, 14, 18, 24, 25, 26, 32, 36, 40, 54, 59, 68, 69, 70, 72, 76, 77, 79, 84, 87, 88, 91, 92], "detect": [0, 3, 26, 77, 88], "detect_format": 15, "determin": [0, 1, 5, 6, 9, 17, 65, 71, 72, 76, 77, 79, 85, 91, 93, 94], "determinenumpag": 1, "determinist": [75, 88], "detoken": [65, 88, 91], "detokenizedgenerationresultbas": 65, "dev": [61, 62, 88], "devel": [27, 28, 60], "develop": [12, 13, 14, 17, 24, 27, 38, 41, 42, 44, 45, 59, 60, 64, 66, 70, 77, 86, 88, 90], "deviat": 69, "devic": [0, 1, 2, 47, 65, 72, 77, 79, 81, 82, 87], "device_id": 82, "device_map": 81, "device_memory_size_v2": 84, "device_request_typ": 79, "deviceallocationnvl": 1, "devicecach": 1, "devicecacheperc": 0, "deviceid": [0, 1, 2], "dgx": [6, 14, 18], "diagon": 77, "diagram": 10, "diamond": 24, "dict": [12, 15, 17, 65, 77, 79, 82, 88, 90, 93], "dict_kei": 87, "dictat": 74, "dictionari": [13, 15, 65, 78], "didn": 74, "differ": [0, 1, 2, 4, 5, 6, 8, 12, 13, 14, 15, 17, 18, 23, 25, 30, 56, 60, 64, 65, 66, 68, 70, 72, 74, 75, 77, 79, 82, 84, 85, 88, 92], "differenti": 77, "difftyp": 1, "diffus": [30, 56, 78, 88], "diffusersattent": 78, "digit": 66, "dilat": [77, 78], "dim": [0, 1, 77, 78, 79, 82, 87], "dim0": 77, "dim1": 77, "dim_head": 78, "dim_in": 78, "dim_out": 78, "dim_rang": 77, "dimems": 1, "dimens": [0, 1, 5, 6, 9, 77, 78, 79, 84, 87, 88, 90], "dimension": 77, "dimrang": 77, "dimtype64": [0, 1], "dir": [32, 60, 64], "direct": [0, 2, 17, 62, 87], "directli": [0, 2, 6, 7, 10, 14, 17, 28, 32, 60, 64, 68, 75, 76, 77, 83, 88, 92, 94], "directori": [3, 12, 13, 14, 15, 17, 25, 50, 51, 52, 60, 65, 68, 69, 70, 79, 82, 83, 88, 90], "disabl": [0, 1, 5, 6, 8, 11, 15, 25, 68, 72, 75, 76, 77, 80, 82, 84, 88], "disable_forward_chunk": 79, "disable_kv_cach": 82, "disable_weight_only_quant_plugin": 79, "disable_xqa": 5, "disablelookahead": 1, "disablelookaheaddecod": 1, "disableseamlesslookaheaddecod": 1, "disadvantag": [17, 71], "disagg_executor": 0, "disaggexecutororchestr": [0, 2], "disaggreg": [0, 59, 65, 88], "disaggregated_param": 65, "disaggregatedparam": 65, "disaggserverbenchmark": [2, 88], "disaggserverutil": 2, "discard": 72, "disclaim": [70, 72, 74, 75], "disclosur": 88, "disconnect": 88, "discourag": [0, 6, 65], "discov": [14, 62], "discrep": [60, 90], "discuss": [5, 70, 72, 75, 76, 88], "disk": [3, 17, 41, 44, 60, 64], "dispatch": [0, 4, 17, 24, 32], "displai": 65, "disservingrequeststat": 0, "disservingstat": 0, "dist": [18, 51, 62, 67, 68, 69, 70], "distanc": [5, 77], "distil": 88, "distinct": [9, 10, 24, 77], "distinguish": 8, "distribut": [1, 4, 5, 6, 14, 24, 36, 37, 68, 77, 82, 84], "distserv": 2, "disturb": 46, "dit": [79, 88], "div": 77, "dive": [66, 67], "divers": [0, 6, 67], "diversity_penalti": 6, "divid": [15, 77, 88], "divup": 77, "dl": 23, "do": [1, 2, 7, 15, 17, 18, 23, 24, 32, 59, 62, 70, 72, 75, 77, 83, 87, 90, 92], "do_cross_attent": [77, 78], "do_layer_norm_befor": 13, "do_sampl": 6, "doc": [1, 18, 22, 24, 28, 72, 75, 77, 87, 88], "docker": [18, 50, 51, 52, 59, 83, 87, 88], "docker_run_arg": 18, "dockerfil": [27, 60], "document": [0, 2, 5, 6, 8, 9, 10, 12, 13, 14, 16, 17, 20, 21, 23, 29, 30, 31, 32, 33, 34, 35, 40, 55, 56, 57, 60, 62, 63, 67, 69, 70, 76, 77, 84, 85, 87, 91, 92], "doe": [0, 2, 5, 9, 10, 17, 18, 19, 25, 68, 69, 75, 77, 82, 84, 86, 88, 90, 94], "doesn": [1, 5, 24, 27, 32, 68, 74, 75], "dollar": 68, "domin": [24, 88], "don": [10, 17, 27, 71, 75, 77], "done": [1, 8, 14, 18, 66, 68, 72, 74, 77, 80, 90], "dongjiyingdji": 88, "dora": [25, 77, 78], "dora_plugin": [9, 25, 77], "dot": [15, 24, 77], "doubl": [0, 20, 73, 75, 87], "down": [0, 2, 3, 9, 19, 53, 66, 71, 77, 82], "down_proj": 15, "download": [16, 50, 51, 52, 53, 60, 61, 62, 64, 68, 70, 83, 87, 88], "downscale_freq_shift": 78, "downsid": 75, "downstream": 85, "dp": [18, 19, 22, 24, 88], "dp8": 24, "dprank": 0, "dpsize": 0, "dq": 59, "draft": [0, 1, 24, 25, 59, 82, 88], "draft_indic": 79, "draft_len": 79, "draft_path": 82, "draft_prob": 79, "draft_target_model": 10, "draft_token": [65, 79], "draft_tokens_extern": [25, 79], "draftacceptancethreshold": 1, "draftbuff": 1, "drafter": 10, "draftindic": 1, "draftlen": 1, "draftlogit": 1, "draftparticipantid": 0, "draftpath": 1, "draftpathshost": 1, "draftprob": 1, "draftrequestid": 0, "drafttoken": [0, 1], "drafttokenid": 1, "drafttokensextern": 1, "dram": 14, "dreamgenx": 88, "drive": [14, 68], "driven": 66, "driver": [84, 88], "drop": [72, 74, 76], "dropout": 78, "dropout_prob": 78, "dry_run": [25, 65, 88], "dst": 1, "dstate": 77, "dsttype": 1, "dt_proj": 77, "dt_rank": 77, "dtype": [1, 7, 9, 11, 12, 13, 14, 17, 65, 68, 70, 71, 77, 78, 79, 80, 81, 82, 87, 88, 93], "dual": 60, "due": [0, 10, 17, 21, 24, 60, 68, 70, 74, 76, 82, 88, 92], "dummi": [65, 70, 88], "dump": [0, 3, 60, 65], "dump_debug_buff": 82, "duplic": 88, "duplicate_data": 77, "durat": [0, 70], "duration_m": 65, "durationm": 0, "dure": [0, 1, 5, 6, 7, 10, 11, 14, 22, 24, 25, 60, 65, 67, 68, 75, 76, 82, 84, 87, 92, 93], "dynam": [0, 24, 25, 65, 68, 77, 79, 82, 84, 88, 94], "dynamic_batch_config": 65, "dynamic_batch_moving_average_window": 65, "dynamic_quant_bf16tonvfp4": 24, "dynamic_tree_max_topk": [39, 65], "dynamicbatchconfig": [0, 65], "dynamicbatchmovingaveragewindow": 0, "dynamicbatchsizeconfig": 0, "dynamicdecodelay": 1, "dynamicqu": 24, "dynamictreemaxtopk": 0, "dynamictreemaxtopkhost": 1, "e": [0, 2, 3, 5, 8, 9, 15, 26, 27, 47, 50, 51, 52, 60, 65, 67, 77, 80, 82, 85, 87, 88, 90], "e2": 59, "e4m3": 20, "e5m2": 20, "each": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 18, 24, 25, 26, 32, 47, 50, 51, 52, 65, 68, 69, 70, 71, 74, 75, 76, 77, 78, 80, 82, 84, 85, 87, 88, 91, 92, 93, 94], "eager": [66, 88], "eagl": [0, 1, 25, 36, 37, 59, 65, 79, 82, 88], "eagle_choic": [39, 65, 82], "eagle_dynamic_tree_max_top_k": 82, "eagle_posterior_threshold": 82, "eagle_temperatur": 79, "eagle_use_dynamic_tre": 82, "eaglechoic": [0, 1], "eagleconfig": [0, 1, 79], "eagledecodingconfig": [39, 65], "eagleforcausallm": 79, "eagleinput": 1, "eaglelastinput": 1, "eaglenetctxcontextlengthshost": 1, "eaglenetctxpastkeyvaluelengthshost": 1, "eaglenetctxrequesttypeshost": 1, "eaglenetgencontextlengthshost": 1, "eaglenetgenpastkeyvaluelengthshost": 1, "eaglenetgenrequesttypeshost": 1, "earli": [82, 87, 88], "earlier": [0, 13, 72, 87], "early_stop": [6, 65, 82, 88], "early_stop_criteria": 82, "earlystop": [0, 1, 6], "eas": [16, 66, 69], "easi": [23, 70], "easier": [14, 17, 18, 68], "easili": [15, 16, 18, 24, 66, 77], "east": [12, 14, 87], "eastern": 83, "ebnf": [0, 3, 65], "echo": [26, 27, 28, 51, 52], "eddi": 88, "edg": 20, "edit": [10, 60], "ef648e7489c040679d87ed12db5d3214": 83, "effect": [0, 2, 6, 10, 24, 25, 62, 65, 72, 74, 75], "effici": [4, 5, 6, 8, 10, 14, 16, 24, 25, 30, 38, 41, 42, 44, 45, 56, 84, 86, 89, 91, 92, 93], "effort": [10, 13, 54, 72, 88], "eg": 69, "eight": [18, 19], "einop": 77, "einstein": 77, "einsum": 77, "einsum_eq": 77, "either": [0, 1, 2, 3, 16, 24, 41, 44, 54, 65, 77, 84, 87, 88], "element": [0, 1, 5, 6, 9, 65, 77, 78, 85], "element_typ": 1, "elementwis": [7, 77], "elementwise_affin": 78, "elementwise_binari": 77, "elementwise_sub": 7, "elementwise_sum": 7, "elementwiseoper": [7, 77], "eleutherai": 68, "elif": 94, "elimin": [2, 10, 24, 25, 66, 68, 72, 74, 88], "ellipsi": 77, "els": [0, 14, 15, 17, 32, 47, 49, 54, 77, 87, 94], "elsinor": 46, "emb": [14, 56, 78], "embark": 66, "embed": [0, 8, 12, 25, 65, 68, 77, 82, 88, 90, 92], "embed_dim": 78, "embed_posit": 78, "embed_positions_for_gpt_attent": 78, "embed_positions_for_gpt_attention_loc": 78, "embed_positions_loc": 78, "embed_token": [15, 90], "embedding_bia": 65, "embedding_dim": 78, "embedding_multipli": 79, "embedding_parallel_mod": 65, "embedding_scal": 79, "embedding_sharding_dim": [13, 79], "embeddingbia": [0, 1], "embeddingt": [0, 1], "emerg": [23, 24], "emphasi": 13, "emploi": [10, 91, 94], "empow": 24, "empti": [0, 1, 10, 32, 77, 88, 94], "emptybuff": 1, "emptygenslot": 0, "emptytensor": 1, "emul": [77, 88], "en": 88, "enabl": [0, 2, 3, 5, 6, 7, 9, 10, 11, 14, 15, 20, 21, 22, 23, 24, 25, 26, 28, 32, 38, 43, 45, 59, 60, 61, 62, 65, 68, 70, 74, 76, 77, 78, 79, 80, 82, 83, 85, 87, 88, 90, 92, 93], "enable_allreduc": 77, "enable_attention_dp": [18, 26, 51, 65], "enable_batch_size_tun": 65, "enable_block_reus": [26, 39, 46, 49, 65], "enable_build_cach": [65, 88], "enable_chunked_context": [82, 88], "enable_chunked_prefil": [65, 88], "enable_context_fmha_fp32_acc": [65, 82], "enable_debug_output": [25, 65, 87], "enable_forward_chunk": 79, "enable_fp8": 54, "enable_if_t": 1, "enable_iter_perf_stat": 26, "enable_kv_cache_reus": 8, "enable_lora": [53, 65], "enable_max_num_tokens_tun": [65, 88], "enable_multi_devic": 88, "enable_nvfp4": 54, "enable_overlap_schedul": [26, 69], "enable_partial_reus": 65, "enable_prompt_adapt": [65, 88], "enable_qkv": 78, "enable_tqdm": 65, "enable_trt_overlap": 88, "enable_ucx": 88, "enable_xqa": 88, "enableattentiondp": [0, 1], "enablebatchsizetun": 0, "enableblockreus": [0, 8], "enablechunkedcontext": 0, "enablecontextfmhafp32acc": 0, "enabled_with_fp32_acc": 5, "enablelookaheaddecod": 1, "enablemaxnumtokenstun": 0, "enablepartialreus": 0, "enableseamlesslookaheaddecod": [0, 1], "enabletrtoverlap": 0, "enc": [25, 82, 88], "enc_dec": 6, "encapsul": [5, 6, 14, 77], "encdecmodelrunn": 82, "encod": [0, 5, 6, 20, 24, 25, 65, 77, 82, 85, 86, 88], "encode_base64_content_from_url": 56, "encoded_vocab": [0, 3], "encodedvocab": [0, 3], "encoder_hidden_st": [78, 79], "encoder_input_featur": 82, "encoder_input_id": 82, "encoder_input_len_rang": 88, "encoder_input_length": [77, 78, 82], "encoder_language_adapter_rout": 82, "encoder_max_input_length": [78, 82], "encoder_output": [78, 79, 82], "encoder_output_length": 82, "encoder_run": 82, "encoderenginebuff": 0, "encoderhiddens": 1, "encoderinputfeatur": 0, "encoderinputtokenid": 0, "encoderjsonconfigstr": 0, "encoderlen": 0, "encodermodel": [0, 79], "encodermodelpath": 0, "encoderoutput": 0, "encoderoutputlength": 0, "encount": [15, 18, 62, 87], "encourag": [0, 6, 17, 65], "end": [0, 1, 5, 6, 14, 25, 39, 44, 48, 49, 54, 65, 66, 68, 72, 75, 76, 77, 83, 88, 93], "end_dim": 77, "end_id": [65, 82, 88], "end_token": [0, 65], "endeavor": 24, "endid": [0, 1], "endpoint": [34, 35, 65, 83, 88], "endswith": 15, "enforc": [70, 77], "engin": [0, 1, 2, 3, 5, 6, 7, 9, 10, 11, 16, 17, 22, 24, 25, 26, 32, 41, 44, 53, 59, 62, 65, 69, 71, 72, 74, 75, 76, 77, 79, 82, 84, 87, 88], "engine_buff": 82, "engine_dir": [11, 12, 13, 14, 17, 65, 68, 70, 82, 83, 87], "engine_inspector": 82, "engine_llama_3": 14, "engine_nam": 82, "engine_output": 25, "engineaddr": 1, "enginebuff": [0, 1], "enginefilenam": 1, "engineinput": 1, "engineoutput": 1, "enginepath": 1, "engines": 1, "enhanc": [4, 6, 10, 18, 24, 66, 76, 84, 89, 92], "enjoi": [28, 38, 41, 42, 44, 45, 47], "enough": [5, 8, 18, 74, 84, 91, 94], "enqueu": [0, 3, 14, 82, 84, 88], "enqueuecontext": 0, "enqueuegener": 0, "enqueuerequest": [0, 2, 3], "ensur": [2, 3, 4, 7, 17, 60, 68, 74, 80, 90, 93], "enter": [7, 27, 69, 74, 93], "enterpris": 40, "entir": [0, 3, 9, 14, 19, 24, 66, 68, 69, 77, 84, 93], "entri": [0, 9, 36, 45, 61, 62, 68, 77, 83, 88], "entrypoint": [27, 64, 70], "enum": [0, 1, 2], "enumer": [0, 1, 43, 47, 89], "env": [26, 29, 30, 31, 33, 34, 35, 68], "envelop": 48, "environ": [6, 10, 18, 24, 30, 50, 51, 52, 56, 59, 60, 62, 67, 68, 70, 72, 74, 75, 87, 88, 89, 92], "environment": 15, "eo": [6, 65], "eof": [18, 26, 51], "eos_token_id": [3, 82], "ep": [4, 18, 24, 26, 68, 77, 78], "ep2": 24, "ep2tp4": 24, "ep4tp2": 24, "ep8tp8": 24, "ep_siz": [26, 33], "epsilon": [0, 77], "eq": 77, "equal": [0, 1, 3, 4, 25, 32, 71, 77, 78, 84], "equal_progress": [65, 76], "equat": [22, 77], "equip": [2, 16], "equival": [24, 72, 77, 90], "equvili": 25, "erenup": 88, "err": [50, 51, 52], "error": [0, 2, 3, 9, 17, 25, 26, 54, 59, 60, 62, 65, 70, 74, 84, 88], "errorcod": 64, "errormsg": 0, "especi": [7, 25, 38, 41, 42, 44, 45, 47, 71, 74, 93], "essenti": [10, 68], "estim": [54, 68, 88, 94], "et": 19, "etc": [0, 1, 10, 62, 67, 72, 75, 82, 84, 87, 90], "ethnzhng": 88, "eval": 40, "evalu": [20, 21, 59, 88], "even": [2, 5, 6, 14, 17, 23, 24, 25, 46, 70, 74, 77, 84], "evenli": [4, 24], "event": [0, 1, 36, 37, 65], "event_buffer_max_s": [46, 65], "event_id": 46, "eventbuffermaxs": 0, "eventid": 0, "eventptr": 1, "ever": [0, 75], "everi": [0, 3, 15, 24, 68, 70, 71, 77, 82], "everyth": 14, "evict": [0, 1, 8, 9, 66, 68, 70, 74], "evolv": [5, 17, 24, 66, 85, 93], "ex": [51, 52], "exact": [5, 84], "exam": 24, "examin": 10, "exampl": [0, 5, 6, 7, 8, 10, 11, 12, 16, 17, 19, 21, 23, 26, 32, 40, 47, 50, 54, 59, 60, 64, 65, 69, 70, 71, 72, 73, 74, 75, 76, 77, 82, 83, 84, 85, 86, 87, 88, 89, 90, 92, 94], "example_logits_processor": 47, "exaon": [15, 86, 88], "exc": 43, "exce": [0, 2, 76, 77], "exceed": [0, 84], "except": [0, 3, 5, 6, 17, 24, 25, 49, 71, 77, 87, 88], "excess": 5, "exchang": 65, "excit": [38, 41, 42, 43, 44, 45], "exclud": [65, 72, 77, 88], "exclude_input_from_output": 65, "exclude_modul": [13, 65, 88], "excludeinputfromoutput": 0, "exclus": [1, 6, 85, 88], "exec": 67, "execut": [0, 2, 3, 6, 9, 10, 14, 16, 17, 24, 59, 65, 66, 67, 68, 74, 76, 77, 82, 83, 84, 91, 94], "executor": [1, 2, 8, 10, 11, 16, 32, 47, 53, 59, 65, 66, 68, 76, 82, 84, 88, 91], "executor_config": 93, "executorconfig": [0, 3, 11], "executorexampledisaggreg": 2, "executorexamplefastlogit": 88, "exhaust": [0, 16], "exist": [1, 6, 8, 9, 10, 15, 17, 24, 25, 46, 62, 65, 68, 82, 88, 92], "exit": [69, 82], "exp": 77, "expand": [0, 21, 23, 77, 82, 88], "expand_dim": 77, "expand_dims_lik": 77, "expand_mask": 77, "expand_shap": 77, "expans": 77, "expect": [0, 5, 6, 12, 14, 15, 17, 21, 25, 32, 50, 51, 52, 59, 65, 68, 70, 73, 77, 87, 88], "expens": [3, 10, 66, 71, 72, 76], "experi": [10, 22, 23, 24, 64, 66, 67, 68, 87], "experiment": [5, 6, 10, 15, 26, 50, 51, 52, 59, 68, 85, 88, 89], "expert": [9, 18, 26, 45, 59, 65, 75, 88], "expertis": 24, "expir": 0, "explain": [6, 14, 16, 74, 77, 84, 85, 91, 92], "explan": [18, 75, 82, 84], "explicit": [0, 1, 10, 77, 88], "explicit_draft_token": [10, 25, 79], "explicitdrafttoken": [0, 1], "explicitdrafttokensinput": 1, "explicitdrafttokenslastinput": 1, "explicitdrafttokensmodul": 1, "expliciteosstop": 0, "explicitli": [1, 2, 7, 10, 14, 15, 25, 26, 32, 65, 88], "explor": [10, 24, 66], "expon": 20, "exponenti": 10, "export": [2, 13, 17, 18, 24, 25, 26, 34, 35, 50, 51, 52, 68, 81, 82, 87, 88], "export_fmt": 89, "expos": [0, 6, 14, 28, 60, 72, 88], "express": [0, 3, 65, 77], "extend": [0, 3, 8, 14, 24, 65, 75, 77, 88], "extended_runtime_perf_knob_config": [65, 88], "extendedruntimeperfknobconfig": [0, 65], "extens": [13, 16, 62, 66, 68, 88], "extern": [0, 7, 15, 82, 84], "external_checkpoint_dir": 15, "external_kei": 15, "external_weight": 15, "externaldrafttoken": 0, "externaldrafttokensconfig": [0, 1], "externaldrafttokensinput": 1, "externalstream": 47, "extra": [0, 2, 5, 8, 10, 13, 18, 24, 25, 26, 33, 62, 69, 71, 72, 82, 88], "extra_arg": 51, "extra_id": 8, "extra_llm_api_opt": [18, 26, 33, 51, 68, 69], "extra_token": 78, "extract": [0, 3, 60, 67, 73, 77, 82], "extrapol": 77, "extrem": [14, 24, 72, 74, 75], "f": [0, 5, 6, 27, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 61, 62, 65, 67, 70, 76, 77, 83, 87, 89], "face": [3, 9, 11, 16, 17, 32, 65, 68, 79, 83, 88], "facilit": [7, 10, 83], "fact": [66, 68, 75], "factor": [23, 71, 72, 77, 78, 84, 85], "factori": [17, 65, 82, 88], "factual": 6, "fail": [65, 82, 84, 87, 94], "failur": [15, 88], "fairli": 14, "fairseq": [86, 88], "fake": [8, 88], "fakebuff": 1, "falcon": [13, 23, 64, 68, 85, 86, 88], "falconconfig": 79, "falconforcausallm": 79, "falconmodel": 79, "fall": [62, 69, 88], "fallback": 15, "fals": [0, 1, 2, 3, 5, 6, 7, 8, 13, 24, 25, 26, 40, 46, 49, 51, 65, 77, 78, 79, 80, 81, 82, 88], "false_output_valu": 77, "false_valu": 77, "famili": [5, 15, 86, 88], "familiar": [6, 14, 64, 70, 71, 73, 83], "famou": [6, 54], "faq": 59, "far": [0, 3], "fast": [0, 5, 10, 65, 68, 71, 88], "fast_build": [25, 65, 88], "fastapi": 88, "fastapi_serv": 88, "faster": [5, 17, 20, 21, 25, 69, 70, 77], "fasterdecod": 49, "fastlogit": 0, "fault": 88, "favor": 88, "favorit": 53, "fc": [13, 14, 15, 87], "fc_gate": 78, "fc_gate_dora": 78, "fc_gate_lora": 78, "fc_gate_plugin": 78, "featur": [0, 2, 3, 5, 7, 9, 10, 13, 14, 15, 17, 23, 24, 25, 50, 51, 52, 59, 60, 68, 72, 74, 75, 76, 77, 80, 82, 86, 89, 90, 92], "feature_dim": 82, "fed": [69, 79], "feed": 77, "feedback": 88, "feedforward": 4, "feel": 53, "fetch": [0, 26, 91], "few": [8, 14, 17, 23, 74], "fewer": [5, 10, 19, 92], "ffn": [4, 24], "ffn_hidden_s": 78, "fhma": 88, "field": [0, 6, 13, 17, 26, 28, 32, 65, 66, 68, 72, 79, 80, 85, 88, 92], "figur": 24, "file": [0, 3, 4, 5, 7, 8, 13, 14, 15, 17, 18, 25, 26, 34, 35, 62, 65, 67, 68, 69, 82, 83, 88, 90], "filepath": 1, "filesystem": [0, 1], "fill": [1, 15, 28, 38, 41, 42, 44, 45, 77, 92], "fill_attention_const_params_for_long_rop": 78, "fill_attention_const_params_for_rop": 78, "fill_attention_param": 78, "fill_none_tensor_list": 78, "fill_valu": [47, 77], "fillemptyfieldsfromruntimedefault": 0, "filloper": 77, "filltaskstensor": 1, "filter_medusa_logit": 82, "final": [0, 1, 9, 24, 25, 26, 27, 32, 77, 94], "final_logit_softcap": 79, "final_output_id": 82, "finalize_decod": 82, "find": [18, 72, 77, 87, 88], "find_best_medusa_path": 82, "fine": [10, 18, 68, 75, 78], "finer": 7, "finetun": 24, "finish": [0, 1, 3, 6, 17, 32, 48, 64, 65, 66, 68, 82, 91, 93], "finish_reason": [48, 65, 83, 88], "finishedst": 1, "finishedsum": 1, "finishreason": [0, 1, 88], "first": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 16, 21, 23, 25, 26, 27, 62, 64, 65, 68, 69, 70, 72, 74, 75, 76, 77, 84, 87, 88, 90, 92, 93, 94], "first_come_first_serv": [65, 76], "first_gen_token": 65, "first_lay": 82, "firstgentoken": 0, "firstit": 0, "firstli": [27, 74, 84], "firstscheduledtim": 0, "firsttokentim": 0, "fit": [1, 5, 19, 20, 65, 71, 72, 94], "fitting_request": 94, "fix": [9, 10, 68, 84], "fjosw": 88, "flag": [0, 1, 3, 5, 9, 17, 22, 26, 32, 59, 68, 72, 73, 74, 76, 77, 84, 88], "flags_siz": 1, "flan": [85, 86], "flash": [5, 14], "flashattent": [5, 14, 83], "flashinf": 92, "flashinferattent": 92, "flashmla": 88, "flatten": [1, 9, 22, 77, 78], "flattenedinouts": 1, "flattenn": 1, "flayer": 7, "flayerinfomemo": 7, "flexibl": [10, 17, 24, 32, 60], "flight": [1, 16, 59, 68, 74, 76, 83, 84, 88], "flip": 77, "flip_sin_to_co": 78, "float": [0, 1, 6, 11, 13, 14, 20, 47, 65, 76, 77, 78, 79, 82, 85], "float16": [7, 9, 11, 12, 13, 17, 25, 71, 77, 79, 80, 83, 87], "float2": 77, "float32": [0, 13, 25, 77, 78, 79, 80], "floattensor": 90, "floattyp": [0, 1], "floor_div": 77, "floordiv": 77, "flow": [7, 17, 24, 70, 71, 72, 74, 75, 88, 91, 94], "fly": [5, 77, 85], "fmha": [0, 25, 65, 77, 82, 84, 88], "fmt_dim": 1, "focu": [7, 23, 24, 67], "focus": [10, 68, 72, 73, 88], "fold": 84, "folder": [0, 3, 6, 17, 70, 85, 86, 88], "folder_trt_llm": 14, "follow": [3, 6, 7, 9, 10, 12, 13, 14, 15, 17, 18, 23, 24, 25, 26, 28, 32, 42, 43, 46, 50, 51, 52, 60, 61, 62, 64, 68, 69, 70, 71, 72, 73, 74, 75, 77, 83, 85, 86, 88, 89, 90, 92, 93], "footprint": [5, 19, 84], "for_each_rank": 79, "forc": [5, 24, 68], "force_drop_id": 78, "force_multi_block_mod": 68, "force_nccl_all_reduce_strategi": 88, "force_num_profil": 65, "force_words_id": 6, "forecast": 10, "foretel": 46, "fork": 67, "form": [0, 3, 5, 10, 65, 77, 83], "format": [0, 3, 13, 15, 17, 20, 23, 35, 59, 60, 64, 65, 66, 70, 72, 82, 83, 84, 87, 88, 92], "former": [14, 23, 46], "formula": 77, "forum": 88, "forward": [0, 1, 7, 10, 12, 14, 47, 76, 77, 78, 79, 87, 88, 90, 91, 92, 93, 94], "forward_loop": 68, "forward_with_cfg": 79, "forward_without_cfg": 79, "forwardasync": 1, "forwarddispatch": 1, "forwardsync": 1, "found": [3, 4, 5, 6, 7, 10, 14, 16, 20, 60, 62, 68, 70, 72, 75, 85, 94], "four": [3, 7, 10, 13, 24, 78], "fourth": 3, "fp": [85, 88], "fp16": [5, 9, 11, 13, 15, 19, 20, 23, 25, 59, 68, 72, 75, 77, 83, 86, 87, 88], "fp32": [0, 5, 24, 25, 59, 65, 77, 82, 83, 86, 87, 88], "fp4": [18, 25, 88], "fp8": [17, 18, 19, 21, 22, 23, 24, 25, 41, 49, 54, 59, 65, 68, 73, 75, 77, 80, 84, 86, 88, 89, 92], "fp8_block_scal": 65, "fp8_blockscale_gemm": 88, "fp8_inputs_overrid": 77, "fp8_kv_cach": [5, 85], "fp8_per_channel_per_token": 65, "fp8_qdq": 85, "fp8_rowwise_gemm_plugin": 25, "fp_valu": 5, "fpa_intb": 88, "fraction": [0, 26, 65, 77, 78, 82], "framework": [10, 12, 13, 16, 17, 66, 77, 88], "franc": [12, 14, 36, 38, 39, 41, 42, 43, 44, 45, 49, 54, 61, 62, 70, 76, 83, 87, 89], "free": [0, 1, 9, 14, 15, 26, 66, 74, 78, 79, 82, 84, 93], "free_gpu_memory_fract": [26, 32, 44, 48, 65, 76, 88], "free_resourc": [91, 93], "freed": 68, "freedom": 17, "freegpumemoryfract": [0, 84, 88], "freenumblock": 0, "french": 83, "freq": 77, "frequenc": [68, 78], "frequency_penalti": [65, 82, 88], "frequencypenalti": [0, 1, 6], "frequent": [8, 87], "friend": [0, 1, 68], "friendli": 77, "from": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 21, 23, 24, 25, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 61, 62, 64, 65, 66, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 79, 81, 82, 83, 84, 87, 88, 89, 90, 91, 92, 93, 94], "from_argu": 79, "from_checkpoint": [17, 79], "from_config": 79, "from_dict": [65, 79], "from_dir": 82, "from_engin": 82, "from_hugging_fac": [12, 15, 17, 79], "from_jax": 17, "from_json_fil": [65, 79], "from_kera": 17, "from_meta_ckpt": [17, 79], "from_nemo": [17, 79], "from_pretrain": 79, "from_prun": 79, "from_serialized_engin": 82, "from_str": 77, "fromfil": 14, "full": [0, 4, 5, 6, 8, 9, 10, 20, 21, 26, 66, 67, 68, 71, 77, 82, 83, 84, 87], "full_lik": 47, "fulli": [36, 88], "function": [0, 1, 3, 5, 11, 12, 14, 16, 17, 24, 64, 65, 66, 67, 75, 80, 82, 84, 85, 86, 87, 88, 93, 94], "further": [3, 4, 5, 10, 14, 19, 23, 25, 68, 72, 75, 92], "furthermor": [10, 24, 72], "fuse": [5, 10, 14, 24, 25, 75, 77, 83, 88, 90, 92], "fuse_a": 24, "fuse_fp4_qu": 25, "fuse_qkv_project": 79, "fuseattentionwithbiaspass": 7, "fused_gate_up_dora": 78, "fused_gate_up_lora": 78, "fusedgatedmlp": [77, 78], "fusevalu": 1, "fusion": [7, 25, 59, 66, 74, 84, 85, 88, 92], "fusion_op": 77, "futur": [2, 5, 6, 10, 15, 17, 23, 25, 36, 38, 39, 40, 41, 42, 43, 44, 45, 49, 54, 60, 61, 62, 64, 65, 66, 68, 70, 76, 77, 83, 84, 85, 88, 89], "fuyu": [86, 88], "g": [3, 15, 26, 47, 50, 51, 52, 65, 74, 82, 90], "g1": 74, "g2": 74, "gain": [71, 74], "gamma": 77, "gate": [9, 15, 25, 70, 77, 88], "gate_a": 77, "gate_a_bia": 77, "gate_bia": 77, "gate_proj": 15, "gate_x": 77, "gate_x_bia": 77, "gatedmlp": [77, 78], "gather": [0, 1, 25, 42, 43, 65, 77, 82], "gather_all_token_logit": [25, 88], "gather_context_logit": [25, 65, 79, 82], "gather_dim": [14, 77], "gather_generation_logit": [25, 65, 79, 82], "gather_last_token_logit": 77, "gather_nd": 77, "gather_output": 78, "gathercontext": [0, 88], "gatheredid": 1, "gatherel": 77, "gathergenerationlogit": 0, "gathermod": 77, "gathertre": 1, "gatherv2": 77, "gb": [2, 21, 60, 65, 68], "gb200": 88, "gcc": 60, "gd": 0, "gdrdma": 2, "geforc": 88, "gegelu": 77, "gegelu_limit": 78, "geglu": 77, "gelu": [77, 79], "gelu_pytorch_tanh": 88, "gelu_tanh": 78, "gemm": [7, 25, 74, 77, 83, 84, 88], "gemm_allreduc": 77, "gemm_allreduce_plugin": [25, 82], "gemm_fc1": 24, "gemm_plugin": [9, 11, 13, 14, 25, 68, 72, 75, 78, 83], "gemm_swiglu": 77, "gemm_swiglu_plugin": [25, 72, 80], "gemma": [17, 64, 85, 86, 88], "gemma2": 86, "gemma2_added_field": 79, "gemma2_config": 79, "gemma3": 88, "gemma3_added_field": 79, "gemma3_config": 79, "gemma_added_field": 79, "gemma_config_kwarg": 79, "gemmaconfig": 79, "gemmaforcausallm": 79, "gen": [65, 88], "genai": [23, 26, 58], "genattent": 24, "genenginepath": 0, "gener": [0, 1, 3, 6, 8, 10, 13, 14, 15, 17, 18, 19, 20, 22, 24, 25, 36, 37, 38, 46, 59, 61, 62, 64, 65, 66, 67, 68, 69, 70, 71, 73, 74, 75, 76, 77, 79, 82, 83, 84, 86, 87, 88, 89, 90, 91, 92, 93, 94], "generate_alibi_bias": 77, "generate_alibi_slop": 77, "generate_async": [32, 42, 43, 65, 88], "generate_logn_sc": 77, "generate_tllm_weight": 15, "generated_text": [36, 39, 49, 53, 61, 62, 70, 76, 83, 89], "generatedtokensperenginestep": 1, "generation_complet": 94, "generation_in_progress": 94, "generation_logit": [48, 65, 82], "generation_onli": 65, "generation_phas": 5, "generation_request": 94, "generation_to_complet": 94, "generationexecutor": [2, 88], "generationlength": 1, "generationlengthsdevic": 1, "generationlengthshost": 1, "generationlengthshostcopi": 1, "generationlogit": 0, "generationmixin": 79, "generationrequestid": 2, "generationresult": 65, "generationsequ": 82, "generationsess": [5, 82, 84], "generationstep": 1, "genericprompttuningparam": 1, "genert": 2, "genexecutorconfig": 0, "genidx": 0, "genrequest": 1, "genrespons": 2, "get": [0, 1, 2, 3, 5, 7, 9, 11, 15, 22, 26, 27, 28, 32, 36, 37, 60, 61, 62, 65, 66, 67, 70, 72, 77, 79, 82, 83, 87, 88, 89, 94], "get_1d_sincos_pos_embed_from_grid": 78, "get_2d_sincos_pos_emb": 78, "get_2d_sincos_pos_embed_from_grid": 78, "get_audio_featur": 82, "get_batch_cache_indic": 93, "get_batch_idx": 82, "get_block_offset": 82, "get_buff": 93, "get_comm": 65, "get_config_group": 79, "get_context_phase_param": 65, "get_device_cap": 54, "get_first_past_key_valu": 78, "get_hf_config": 79, "get_input": 7, "get_kv_cache_ev": [46, 65], "get_kv_cache_events_async": 65, "get_max_resource_count": [93, 94], "get_needed_resource_to_complet": [93, 94], "get_next_medusa_token": 82, "get_num_free_block": 93, "get_num_heads_kv": 82, "get_output": [7, 14], "get_par": [7, 77], "get_request_typ": 65, "get_rope_index": 82, "get_seq_idx": 82, "get_shap": 15, "get_slic": 15, "get_stat": [65, 88], "get_stats_async": 65, "get_timestep_embed": 78, "get_us": [7, 77], "get_visual_featur": 82, "get_vocab": [0, 3], "get_weight": 78, "getacceptancethreshold": 0, "getacceptedlengthscumsum": 1, "getacceptedpackedpath": 1, "getadditionalmodeloutput": 0, "getadditionaloutputnam": 0, "getaddress": 1, "getallnewtoken": 1, "getallottedtimem": 0, "getattentionconfig": 0, "getbackend": 0, "getbadword": 0, "getbatchingtyp": 0, "getbatchsizet": 0, "getbeamsearchbuff": 1, "getbeamsearchdiversityr": 0, "getbeamwidth": 0, "getbeamwidtharrai": 0, "getbuffermanag": 1, "getcachest": 0, "getcachetransceiverconfig": 0, "getcapac": 1, "getcapacityschedulerpolici": 0, "getclientid": 0, "getcommptr": 1, "getcommst": 0, "getcommunicationmod": 0, "getcommunicationtyp": 0, "getconfig": 0, "getconnect": 0, "getcontextchunkingpolici": 0, "getcontextexecutor": 0, "getcontextfmha": 1, "getcontextparallel": 1, "getcontextparallelgroup": 1, "getcontextparallelrank": 1, "getcontextphaseparam": 0, "getcopyonpartialreus": 0, "getcpu": 1, "getcpudiff": 1, "getcrossattentionmask": 0, "getcrosskvcachefract": 0, "getcudagraphcaches": 0, "getcudagraphmod": 0, "getcumlogprob": 1, "getdata": 0, "getdatatyp": [0, 1], "getdatatypenam": 1, "getdebugconfig": 0, "getdebuginputtensor": 0, "getdebugoutputtensor": 0, "getdebugtensornam": 0, "getdebugtensorsmaxiter": 0, "getdecodedurationm": 0, "getdecoderetentionprior": 0, "getdecoderst": 1, "getdecoderstream": 1, "getdecodingconfig": 0, "getdecodingmod": 0, "getdefaultbatchslot": 1, "getdefaulteaglechoic": 1, "getdevic": 1, "getdevicecacheperc": 0, "getdeviceid": 0, "getdeviceof": 1, "getdimens": 1, "getdrafttoken": 0, "getdynamicbatchconfig": 0, "getdynamicbatchmovingaveragewindow": 0, "getdynamictreemaxtopk": 0, "geteaglechoic": 0, "geteagleconfig": 0, "getearlystop": 0, "getembeddingbia": 0, "getembeddingt": 0, "getenablebatchsizetun": 0, "getenableblockreus": 0, "getenablechunkedcontext": 0, "getenablecontextfmhafp32acc": 0, "getenablemaxnumtokenstun": 0, "getenablepartialreus": 0, "getenabletrtoverlap": 0, "getencodedvocab": 0, "getencoderhiddens": 1, "getencoderinputfeatur": 0, "getencoderinputtokenid": 0, "getencoderoutputlength": 0, "getendid": 0, "geterrormsg": 0, "geteventbuffermaxs": 0, "getexecutionconfig": 1, "getextendedruntimeperfknobconfig": 0, "getexternaldrafttokensconfig": 0, "getfastlogit": 0, "getfinishedstep": 1, "getfinishedsum": 1, "getfinishreason": 1, "getfirstgentoken": 0, "getfreegpumemoryfract": 0, "getfrequencypenalti": 0, "getgatheredid": 1, "getgathergenerationlogit": 0, "getgemmallreducedtyp": 1, "getgenexecutor": 0, "getgpu": 1, "getgpudiff": 1, "getgpuspergroup": 1, "getgpuspernod": 1, "getgpuweightsperc": [0, 11], "getguid": 0, "getguideddecodingconfig": 0, "getguideddecodingparam": 0, "getguidetyp": 0, "gethiddens": 1, "gethostcaches": 0, "gethostmemori": 1, "getid": 1, "getinittozero": 1, "getinputtokenextraid": 0, "getinputtokenid": 0, "getinst": 1, "getipcunicastpoint": 1, "getisorchestr": 0, "getiterstatsmaxiter": 0, "getjointdecodinginput": 1, "getjointdecodingoutput": 1, "getkvcacheconfig": 0, "getkvcacheconfigref": 0, "getkvcacheeventmanag": 0, "getkvcacheretentionconfig": 0, "getkvcachetyp": 1, "getkvdatatyp": 1, "getlanguageadapteruid": 0, "getlastrank": 1, "getlatestdebugtensor": 0, "getlatestev": 0, "getlatestiterationstat": [0, 3], "getlatestrequeststat": 0, "getlayertyp": 1, "getlengthpenalti": 0, "getlevel": 1, "getlocalrank": 1, "getlogit": 0, "getlogitsdtyp": 1, "getlogitspostprocessor": 0, "getlogitspostprocessorconfig": 0, "getlogitspostprocessornam": 0, "getlogprob": 1, "getlookaheadconfig": 0, "getlookaheaddecodingconfig": 0, "getlookaheaddecodingmaxnumrequest": 0, "getloraconfig": 0, "getloramodul": 1, "getloraprefetchdir": 0, "getmanagedweightsmapopt": 1, "getmanageweightstyp": 1, "getmaxadapters": 0, "getmaxattentionwindowvec": 0, "getmaxbatchs": [0, 1], "getmaxbeamwidth": [0, 1], "getmaxdecodingdecodertoken": 1, "getmaxdecodingdrafttoken": 1, "getmaxdecodingenginetoken": 1, "getmaxdecodingtoken": 1, "getmaxdraftpathlen": 1, "getmaxencoderlen": 1, "getmaxinputlen": 1, "getmaxlorarank": 1, "getmaxnonleafnodesperlay": 1, "getmaxnumpath": 1, "getmaxnumtoken": [0, 1], "getmaxpagesperblock": 1, "getmaxpagesperblockdevic": 0, "getmaxpagesperblockhost": 0, "getmaxpathlen": 1, "getmaxpositionembed": 1, "getmaxpromptembeddingtables": 1, "getmaxqueues": 0, "getmaxseqidlemicrosecond": 0, "getmaxsequencelen": 1, "getmaxsequencelength": 1, "getmaxtoken": 0, "getmedusachoic": [0, 1], "getmemorytyp": [0, 1], "getmemorytypenam": 1, "getminp": 0, "getmintoken": 0, "getmlphiddens": 1, "getmodelconfig": [0, 1], "getmodelconfigmut": 1, "getmodelnam": 1, "getmodelvari": 1, "getmpist": 0, "getmropeconfig": 0, "getmropepositiondelta": 0, "getmroperotarycossin": 0, "getmultiblockmod": 0, "getmulticastpoint": 1, "getmultimodalembed": 0, "getnam": [0, 1], "getnbattentionlay": 1, "getnbhead": 1, "getnbkvhead": 1, "getnblay": 1, "getnbrnnlay": 1, "getnextdrafttoken": 1, "getnextdrafttokenslength": 1, "getngrams": 0, "getnoderank": 1, "getnoderankof": 1, "getnorepeatngrams": 0, "getnormalizelogprob": 0, "getnumcopystream": [0, 1], "getnumdecodingenginetoken": 1, "getnumdevicemodulelay": 0, "getnumensurework": 0, "getnumhostmodulelay": 0, "getnumkvheadsperlay": 1, "getnumkvheadsperlayerlocalrang": 1, "getnumlanguag": 1, "getnumnod": 0, "getnumpackedmask": 1, "getnumpag": 1, "getnumputwork": 0, "getnumresponsesreadi": 0, "getnumreturnbeam": [0, 1], "getnumreturnsequ": 0, "getnumtransformerlay": 1, "getonboardblock": 0, "getoptimaladapters": 0, "getoptprofilessplitpoint": 1, "getorchestratorconfig": 0, "getorchleadercomm": 0, "getoutputconfig": 0, "getpadid": 0, "getpagedcontextfmha": 1, "getpageptr": 1, "getpagewidth": 1, "getparallelconfig": 0, "getparentid": 1, "getparticipantid": 0, "getpath": 1, "getpathopt": 1, "getpeftcacheconfig": 0, "getperblockretentionprioritydur": 0, "getpin": 1, "getpinneddiff": 1, "getpinnedpool": 1, "getpinnedpooldiff": 1, "getpipelineparallel": 1, "getpipelineparallelgroup": 1, "getpipelineparallelrank": 1, "getpositionid": 0, "getposteriorthreshold": 0, "getppreducescatt": 1, "getprecis": 1, "getpresencepenalti": 0, "getprevdrafttokenslength": 1, "getprior": 0, "getprocessorbatch": 0, "getprocessormap": 0, "getprompttableoffload": 0, "getprompttuningconfig": 0, "getquantmod": 1, "getrank": 1, "getrecvpollperiodm": 0, "getrepetitionpenalti": 0, "getrepl": 0, "getreqid": 0, "getrequestid": 0, "getrequeststatsmaxiter": 0, "getrequesttyp": 0, "getresult": [0, 2, 3], "getreturnallgeneratedtoken": 0, "getrnnconfig": 1, "getrotaryembeddingdim": 1, "getruntimedefault": 1, "getruntimetyp": 0, "getsamplingconfig": [0, 1], "getschedulerconfig": 0, "getschedulerconfigref": 0, "getse": 0, "getsecondaryoffloadminprior": 0, "getselfidx": 0, "getsequencelength": 1, "getserializedst": 0, "getshap": [0, 1], "getsinktokenlength": 0, "getsiz": [0, 1], "getsizeinbit": 1, "getsizeinbyt": [0, 1], "getsizeperhead": 1, "getskipcrossattnblock": 0, "getslotsperpag": 1, "getsocketst": 0, "getspawnprocess": 0, "getspecdecconfig": 0, "getspeculativedecodingmod": 1, "getspeculativedecodingmodul": 1, "getspeculativedecodingmoduleptr": 1, "getstat": 0, "getstatu": 1, "getstoptokenid": 0, "getstopword": 0, "getstream": [0, 1], "getsumlocalkvhead": 1, "gettag": 0, "gettaskid": 0, "gettemperatur": 0, "gettensorparallel": 1, "gettensorparallelgroup": 1, "gettensorparallelrank": 1, "getter": 6, "gettoken": 0, "gettokenizerstr": 0, "gettokenrangeretentionconfig": 0, "gettokensperblock": 1, "gettopk": 0, "gettopp": 0, "gettoppdecai": 0, "gettoppmin": 0, "gettoppresetid": 0, "gettotalnumpag": 1, "gettyp": 1, "getunderlyingdecod": 1, "getunicastpoint": 1, "getusegpudirectstorag": 0, "getuvm": 1, "getuvmdiff": 1, "getverificationsets": 0, "getvers": 1, "getvocabs": 1, "getvocabsizepad": 1, "getweight": 0, "getwindows": 0, "getworkerexecutablepath": 0, "getworlds": 1, "gh200": 88, "ghost": 46, "ghz": 40, "gib": [8, 84], "gid": 0, "gigabyt": 21, "git": [9, 18, 60, 64, 83, 87, 89], "github": [17, 18, 24, 60, 64, 66, 83, 88, 89], "give": [3, 66, 72, 74, 79], "given": [0, 1, 3, 6, 9, 15, 17, 21, 64, 65, 67, 73, 74, 77, 78, 79, 81, 82, 84, 85, 88, 93], "givyboi": 53, "glm": [64, 77, 86, 88], "glm4": [64, 88], "global": [0, 5, 14, 24, 88], "global_max_input_length": 82, "global_max_output_length": 82, "globalrequestid": 0, "glossari": [19, 22], "gm": 87, "gnu": 60, "go": [5, 6, 46, 71, 88], "goal": 76, "goe": [64, 68], "good": [3, 14, 18, 68, 71, 74, 75], "got": [0, 38, 40, 41, 42, 43, 44, 45, 46, 47, 53, 54, 64, 68, 87], "gpqa": 24, "gpt": [1, 5, 10, 14, 16, 20, 23, 25, 59, 64, 68, 77, 84, 85, 86, 87, 88], "gpt2": [79, 87], "gpt3": 21, "gpt_attent": [5, 7, 22, 77, 83, 88], "gpt_attention_plugin": [9, 14, 25, 68, 78, 82, 87, 88], "gpt_attention_plugin_remove_pad": 7, "gpt_variant": [79, 88], "gptattent": 7, "gptattentionpluginremovepaddingrewritepass": 7, "gptconfig": 79, "gptdecod": 6, "gptdecoderbatch": 88, "gptdecoderptr": 1, "gptforcausallm": 79, "gptj": 79, "gptjconfig": 79, "gptjforcausallm": 79, "gptjmodel": 79, "gptlmheadmodel": 87, "gptmanag": 88, "gptmanagerbenchmark": [8, 60, 88], "gptmodel": 79, "gptmodelconfig": 88, "gptneoxforcausallm": 79, "gptneoxmodel": 79, "gptq": [23, 59, 86, 88], "gptsession": 88, "gptsessionbenchmark": 88, "gpu": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 13, 16, 17, 20, 21, 22, 23, 25, 26, 32, 50, 51, 52, 54, 59, 60, 62, 64, 65, 69, 70, 71, 72, 75, 77, 79, 82, 83, 86, 87, 88, 91, 92], "gpu_weights_perc": [11, 82], "gpumemusag": [0, 26], "gpus_per_nod": [25, 26, 65], "gpuspernod": [1, 6], "gpusync": 1, "gpuweightsperc": [0, 11], "gqa": [5, 19, 22, 25, 77, 88, 92], "grace": [8, 59, 86], "gradient": 20, "gradual": 17, "grain": 7, "gram": 10, "grammar": [0, 3, 65], "granit": [86, 88], "graph": [0, 14, 18, 59, 65, 67, 68, 77, 82, 83, 84, 87, 88, 92, 93], "graph_rewrit": 7, "graphic": 48, "gre": 26, "great": [19, 48], "greater": [0, 2, 5, 22, 23, 24, 25, 77], "greatli": [8, 17, 72, 75], "greedi": [0, 6, 91], "greedy_sampl": [39, 65], "greedysampl": 0, "greedysamplinghost": 1, "grid": [14, 72, 74, 77, 78], "grid_search_engin": 70, "grid_siz": 78, "grok": [86, 88], "ground": 67, "groundbreak": 66, "group": [0, 3, 4, 6, 14, 19, 59, 65, 77, 78, 85, 88, 92], "group_cl": 79, "group_norm": 77, "group_siz": [13, 65, 77], "groupedrmsnorm": 24, "groupnorm": [77, 78], "grow": [1, 10, 74], "gt": 77, "gtc": [18, 24], "guarante": [0, 6, 8, 17, 68, 69, 70, 72, 76], "guaranteed_no_evict": [0, 65, 68, 76], "guaranteednoevictschedul": 94, "guard": [46, 70], "guid": [0, 14, 18, 23, 36, 37, 59, 64, 65, 66, 67, 69, 70, 71, 72, 75, 77, 87, 88, 92], "guidanc": [10, 26, 75, 78, 79], "guided_decod": [40, 65], "guided_decoding_backend": [40, 65], "guideddecodingbackend": 0, "guideddecodingconfig": [0, 3], "guideddecodingparam": [0, 3, 40, 65], "guidelin": [2, 71], "guidetyp": [0, 3], "gw": 7, "h": [2, 3, 5, 10, 15, 25, 26, 29, 30, 31, 70, 77, 79, 83, 88], "h1": 77, "h100": [17, 23, 25, 66, 69, 70, 72, 73, 74, 88], "h20": 25, "h200": [20, 25, 69, 88], "h2d": 47, "ha": [0, 1, 3, 5, 8, 9, 13, 14, 15, 17, 18, 19, 23, 24, 25, 28, 60, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 82, 84, 85, 87, 88, 91, 93, 94], "had": [17, 72, 74], "half": [0, 1, 14, 70, 77], "halv": [20, 77], "hand": [8, 10, 16, 71], "handl": [1, 2, 4, 15, 17, 19, 24, 70, 72, 74, 75, 76, 77, 78, 90, 91], "handle_per_step": 82, "hang": [0, 64, 87, 88], "happen": [3, 6, 8, 14, 62, 84, 87], "happi": 82, "hard": 5, "harder": 6, "hardwar": [23, 32, 59, 60, 88], "has_affin": 77, "has_bia": 77, "has_config_group": 79, "has_position_embed": 82, "has_scal": 77, "has_token_type_embed": 82, "has_zero_point": [13, 65], "hascontextawaitthread": 0, "hasdraftlogit": 1, "haserror": [0, 3], "hasgenawaitthread": 0, "hash": [0, 65], "hasresult": 0, "hasrnnconfig": 1, "hasspeculativedecodingmodul": 1, "hattizai": 88, "have": [0, 1, 3, 4, 5, 6, 8, 9, 10, 13, 14, 15, 17, 18, 19, 21, 23, 24, 25, 27, 46, 49, 50, 51, 52, 53, 62, 64, 65, 66, 67, 68, 70, 71, 72, 73, 74, 75, 76, 77, 82, 83, 84, 86, 87, 88, 90], "hbm3": 69, "hbm3e": 21, "he": 46, "head": [1, 6, 10, 14, 19, 25, 49, 54, 59, 68, 77, 78, 88, 92], "head_dim": [92, 93], "head_siz": [5, 77, 79, 82, 88], "header": 2, "headsiz": 77, "headsperlay": 1, "health": [26, 53], "heat": 6, "heavi": 75, "heavier": 71, "height": [35, 78, 82], "hello": [36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 50, 53, 54, 61, 62, 70, 76, 83, 89], "help": [2, 3, 5, 7, 14, 24, 25, 26, 29, 30, 40, 47, 49, 55, 56, 60, 67, 68, 69, 70, 73, 74, 75, 76, 77, 83, 88, 91], "helper": [1, 77], "henc": 90, "here": [2, 3, 7, 9, 11, 12, 13, 14, 15, 17, 18, 20, 21, 26, 28, 32, 36, 40, 60, 67, 70, 71, 72, 74, 75, 77, 82, 83, 84, 85, 87, 89, 92, 93, 94], "heterogen": 2, "heurist": [5, 68, 77, 88], "hf": [6, 9, 11, 15, 25, 26, 41, 42, 43, 44, 45, 49, 50, 51, 52, 54, 68, 69, 70, 82, 86, 87, 89], "hf_config_or_dir": 79, "hf_lora_convert": 9, "hf_model": [68, 79], "hf_model_dir": [11, 12, 13, 17, 79], "hf_model_nam": 68, "hf_model_or_dir": 79, "hf_quant_config": 68, "hf_token": 68, "hfconfigordir": 79, "hgx": 21, "hi": 9, "hidden": [0, 3, 4, 5, 6, 9, 10, 24, 65, 77, 78, 88], "hidden_act": [13, 78, 79], "hidden_dim": [0, 5, 77], "hidden_dim_per_head": [5, 77], "hidden_dtyp": 78, "hidden_s": [0, 7, 13, 15, 77, 78, 79, 82, 90, 92], "hidden_size_in": 9, "hidden_size_out": 9, "hidden_size_per_head": 77, "hidden_st": [12, 77, 78, 79, 82, 87, 90], "hidden_states_for_emb": 79, "hiddens": [0, 1, 6], "hide": 24, "hierarch": 13, "hierarchi": [17, 77], "high": [3, 10, 12, 14, 17, 19, 23, 24, 64, 68, 76, 77, 84, 88], "higher": [0, 1, 5, 6, 8, 9, 10, 15, 19, 20, 22, 66, 69, 76, 84, 88, 90], "highest": [6, 7, 20, 21], "highli": [10, 14, 67, 72], "highlight": [20, 23, 72, 74], "himself": 46, "hint": [68, 77], "hit": [0, 69, 74, 75, 88], "hk": 10, "ho": 9, "hoc": [17, 82], "hold": [0, 1, 3, 4, 7, 8, 9, 10, 65, 71, 78, 84, 91], "home": [18, 54, 68], "homo_head_pattern": 78, "homogen": 2, "hope": 24, "hopper": [5, 8, 18, 19, 20, 23, 25, 59, 60, 66, 72, 86, 88], "horatio": 46, "horizont": 25, "host": [1, 9, 26, 28, 33, 47, 52, 59, 60, 65, 75, 77, 88], "host_cache_s": 65, "host_context_length": [77, 78, 79, 82, 87], "host_context_progress": [77, 78, 87], "host_cross_kv_cache_block_offset": [78, 82], "host_cross_kv_cache_pool_map": 78, "host_cross_kv_cache_pool_point": 78, "host_kv_cache_block_offset": [77, 78, 82, 87], "host_kv_cache_block_point": 87, "host_kv_cache_pool_map": [77, 78, 87], "host_kv_cache_pool_point": [77, 78, 87], "host_max_attention_window_s": [77, 78, 87], "host_past_key_value_length": [77, 78, 87], "host_request_typ": [77, 78, 79, 87], "host_runtime_perf_knob": [77, 78, 87], "host_sink_token_length": [77, 78, 87], "hostcaches": [0, 8], "hostmemori": 1, "hostnam": 26, "hour": 70, "hous": 71, "how": [0, 2, 3, 10, 12, 14, 15, 17, 25, 28, 36, 50, 59, 64, 65, 67, 70, 72, 73, 75, 77, 83, 84, 85, 87, 89, 91, 92], "howev": [2, 3, 5, 10, 17, 18, 19, 24, 26, 68, 71, 72, 74, 75, 76, 84, 88, 90, 91], "hpc": 20, "html": [1, 77, 87], "http": [0, 1, 4, 9, 17, 18, 24, 25, 26, 29, 30, 31, 55, 56, 57, 60, 61, 62, 64, 77, 83, 85, 87, 88, 89], "hub": [16, 53, 65, 68, 83, 88, 89], "hug": [3, 9, 11, 16, 17, 32, 65, 68, 79, 83, 88], "huggingfac": [0, 9, 12, 13, 15, 17, 18, 30, 53, 56, 64, 68, 69, 70, 83, 86, 87, 88, 90], "huggingface_exampl": 89, "huggingface_hub": 53, "huggingface_model_card": 89, "human": [24, 68], "hurt": 75, "hw": 24, "hybrid": [4, 88], "hyper": 13, "hypothesi": 10, "i": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 22, 23, 24, 25, 26, 28, 29, 31, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 60, 61, 62, 64, 65, 66, 68, 69, 70, 72, 73, 74, 75, 76, 77, 78, 79, 80, 82, 83, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94], "ia3": 5, "iactivationlay": 14, "ibrahimamin1": 88, "ibufferptr": 1, "iconstantlay": 77, "icudaengin": [82, 84], "id": [0, 1, 3, 8, 32, 43, 65, 68, 69, 77, 78, 82, 83, 92, 93], "idea": [9, 75], "ideal": [7, 72, 74, 88], "ident": [3, 8, 25, 77], "identifi": [0, 6, 9, 10, 14, 68, 74, 77], "idl": 0, "idtyp": [0, 3], "idx": 82, "ieee": 85, "ieinsumlay": 77, "ielementwiselay": 77, "iexecutioncontext": [82, 84], "ifb": [10, 88], "ifilllay": 77, "igatherlay": 77, "ignor": [25, 65, 68, 77, 82], "ignore_eo": [65, 88], "igptdecod": 1, "ihostmemori": [1, 14, 82], "ii": [5, 77], "ij": 77, "ijk": 77, "ijl": 77, "ik": 77, "ikl": 77, "ilay": [7, 14], "illustr": [7, 10, 16, 24], "ilogg": 1, "ilooplay": 77, "imag": [26, 30, 35, 50, 51, 52, 56, 59, 61, 62, 68, 78, 82, 88], "image64": 56, "image_grid_thw": 82, "image_patches_indic": 82, "image_path": 82, "image_s": 79, "image_token_index": 82, "image_url": [30, 56], "imatrixmultiplylay": 77, "imbal": 74, "immedi": [5, 10, 66, 70, 87], "immut": 1, "impact": [10, 19, 23, 24, 26, 53, 71, 72, 74, 75, 76], "imped": 23, "impl": [0, 94], "implement": [2, 3, 5, 6, 10, 13, 14, 16, 17, 19, 47, 59, 66, 77, 78, 83, 85, 86, 87, 88, 90, 91, 93, 94], "implicit": [1, 5, 10, 77], "implicitli": 1, "import": [10, 15, 17, 19, 23, 26, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 55, 56, 57, 59, 61, 62, 70, 72, 74, 75, 76, 83, 86, 88, 89, 90, 91, 93], "impos": 23, "improv": [5, 8, 14, 19, 20, 21, 22, 23, 24, 25, 38, 41, 42, 44, 45, 59, 66, 68, 69, 70, 72, 73, 74, 75, 88, 89, 92], "in_channel": 78, "in_featur": [13, 14, 78], "in_hidden_s": 77, "in_len": 7, "in_point": 77, "in_progress": 82, "includ": [0, 1, 2, 3, 5, 6, 8, 9, 10, 13, 14, 15, 16, 19, 20, 22, 25, 26, 32, 40, 46, 54, 60, 62, 64, 65, 66, 72, 75, 77, 83, 85, 87, 88, 91, 92, 93, 94], "include_stop_str_in_output": 65, "inclus": 77, "incompat": [25, 88, 89], "incorpor": [0, 24, 66, 88], "incorrect": [8, 10, 88], "increas": [0, 5, 8, 10, 14, 18, 20, 21, 24, 25, 67, 68, 70, 72, 75, 76, 77, 88, 94], "incred": 66, "increment": [60, 88], "incur": [14, 24], "inde": 84, "independ": [0, 1, 2, 3, 10, 77], "index": [0, 1, 3, 10, 15, 24, 32, 48, 59, 61, 62, 65, 77, 82, 83, 88, 92], "index_select": 77, "indic": [0, 1, 3, 5, 6, 10, 13, 65, 76, 77, 78, 82, 84, 93], "indim": 1, "indimfirst": 1, "indirect": 1, "individu": [24, 88], "indivis": 88, "industri": 68, "ineffici": [5, 24], "inetworkdefinit": [7, 14, 77], "inevit": 14, "inf": 47, "infeas": 3, "infer": [0, 2, 6, 9, 10, 14, 16, 17, 18, 19, 20, 21, 24, 25, 30, 56, 59, 64, 67, 69, 70, 71, 72, 73, 75, 76, 77, 82, 85, 87, 88, 91], "infer_shap": 82, "inferencerequest": 88, "infin": 28, "infinit": [14, 68, 69], "inflat": 24, "inflight": [0, 5, 9, 10, 26, 63, 65, 68, 73, 74, 77, 88, 92, 94], "inflight_request_id": 94, "inflightbatch": 0, "inflightbatchingstat": [0, 26], "influenc": [24, 75], "info": [0, 25, 26, 68, 84, 87], "inform": [0, 1, 2, 3, 5, 6, 10, 13, 14, 19, 22, 24, 26, 59, 66, 68, 70, 86, 87, 88], "infti": 6, "inherit": [15, 17, 77, 90, 91, 93, 94], "init": [1, 18, 60, 88], "init_audio_encod": 82, "init_image_encod": 82, "init_llm": 82, "init_processor": 82, "init_token": 82, "initi": [1, 2, 10, 15, 24, 47, 65, 68, 72, 74, 75, 84, 87, 88, 90, 92, 94], "initializer_list": [0, 1], "initmemorypool": 84, "inittozero": 1, "inlin": [0, 1], "inner": 77, "inner_layernorm": [78, 79], "inp": 77, "inpaint": [30, 56], "inprogress": 1, "input": [0, 1, 3, 6, 7, 8, 9, 10, 14, 15, 18, 19, 20, 21, 22, 23, 24, 25, 26, 32, 34, 35, 51, 56, 59, 64, 65, 67, 68, 69, 70, 71, 73, 75, 76, 77, 78, 79, 82, 84, 86, 87, 88, 90, 91, 92, 94], "input_1": 77, "input_1_": 77, "input_audio": 82, "input_featur": 79, "input_fil": 88, "input_id": [8, 12, 24, 68, 77, 79, 82, 87, 90], "input_imag": 82, "input_layernorm": [12, 13, 15, 90], "input_length": [77, 78, 79, 82], "input_list": 77, "input_n": 77, "input_n_": 77, "input_text": [12, 14, 82, 83], "input_timing_cach": [25, 65], "input_token_extra_id": 82, "inputbuff": 1, "inputdesc": 14, "inputdtyp": 1, "inputgentokenshost": 1, "inputlen": 1, "inputpack": [1, 6], "inputs_emb": 90, "inputtokenextraid": 0, "inputtokenid": 0, "insert": [7, 14, 68, 77], "insertinputtensor": 1, "insid": [1, 10, 15, 17, 18, 60, 62, 77, 84, 92], "insight": 24, "insiz": 1, "inspect": [25, 67, 84], "instabl": 2, "instal": [17, 26, 27, 50, 51, 52, 60, 64, 70, 83, 88, 90], "instanc": [0, 2, 3, 6, 7, 10, 14, 24, 32, 47, 64, 65, 82, 84, 88, 92], "instance_idx": 87, "instanti": [70, 76, 93], "instead": [7, 8, 10, 14, 17, 18, 19, 32, 60, 75, 76, 77, 84, 88], "instruct": [10, 18, 26, 30, 35, 41, 56, 60, 68, 69, 70, 71, 75, 76, 83, 86, 88, 89, 90], "int": [0, 1, 6, 12, 13, 14, 17, 43, 47, 65, 74, 77, 78, 79, 82, 90, 92, 93, 94], "int32": [1, 5, 25, 77, 80, 87], "int32_t": [0, 1, 77], "int4": [15, 17, 23, 25, 32, 54, 59, 86, 88], "int4_weight": 85, "int64": [1, 6, 77, 87], "int64_t": [0, 1], "int8": [1, 13, 15, 17, 23, 25, 59, 65, 72, 77, 84, 86, 88], "int8_kv_cach": [5, 85, 88], "int8_t": [0, 1], "int8_weight": 85, "int8awq": 72, "int_clip": 77, "integ": [5, 65, 68, 77, 85, 88], "integr": [10, 88, 91, 92, 93, 94], "intellig": 66, "intend": 84, "intent": 70, "intention": 17, "intenum": 77, "inter": [2, 70, 71, 72, 74, 75, 87, 88], "inter_layernorm": 79, "inter_s": 15, "interact": [3, 10, 66, 83, 87], "interchang": 64, "interconect": 71, "interconnect": [6, 70, 71, 72, 74, 75], "interest": 68, "interfac": [14, 17, 70, 82, 88, 90, 91], "interfer": 87, "interleav": [5, 14], "intermedi": [5, 14, 65, 87], "intermediate_s": [13, 79], "intern": [1, 3, 5, 17, 18, 24, 70, 73, 84, 87, 93], "internal_error": [25, 26], "internlm": [64, 85, 86, 88], "internlm2": [85, 86, 88], "internvl2": 88, "interpol": 77, "interpolation_scal": 78, "interpret": [3, 60, 74], "intersect": 2, "intertwin": 75, "intflag": [79, 81], "intpsplitdim": 1, "intra": 71, "introduc": [17, 20, 24, 28, 85, 88], "introduct": [73, 83, 88], "intuit": [66, 73], "inv": 77, "inv_freq": 77, "invalid": [87, 88], "inventori": 68, "invers": 5, "invest": 68, "investig": [18, 88], "invit": 54, "invoc": 88, "invok": [0, 3, 7, 64, 87, 94], "invokequant": 14, "involv": [0, 1, 2, 10, 14, 23, 78, 91, 92, 93], "io": [5, 27, 28, 84, 88], "ip": [0, 88], "ipc": 60, "ipc_uc_handl": 1, "ipc_uc_ptr": 1, "ipc_uc_va": 1, "ipcmemori": 1, "ipcnvl": 1, "ipcnvlsalloc": 1, "ipcnvlsfre": 1, "ipcnvlshandl": 1, "ipcnvlssupport": 1, "ipluginv3lay": 77, "ireducelay": 77, "irrespect": [0, 6, 47, 65], "is_alibi": 77, "is_caus": 78, "is_const_v": 1, "is_cuda_graph": 92, "is_def": 77, "is_dora": 9, "is_dynam": 77, "is_enc_dec": 82, "is_expert": 78, "is_gated_activ": 77, "is_gemma_2": 79, "is_gemma_3": 79, "is_loc": 78, "is_medusa_mod": 82, "is_mla_en": 77, "is_mla_enabled_flag": 77, "is_module_excluded_from_quant": 65, "is_mrop": 77, "is_network_input": 77, "is_orchestrator_mod": 82, "is_qkv": 78, "is_redrafter_mod": 82, "is_rop": 77, "is_trt_wrapp": 77, "is_valid": 78, "is_valid_cross_attn": 78, "isauto": 0, "isbeamsearch": 0, "iscontextparallel": 1, "iscontinuouskvcach": 1, "iscrossattent": 1, "isdon": 1, "isdora": 1, "isdrafttokensextern": 1, "iseagl": [0, 1], "iselectlay": 77, "isexplicitdrafttoken": [0, 1], "isexternaldrafttoken": 0, "isfin": [0, 3], "isfirstcontextparallelrank": 1, "isfirstpipelineparallelrank": 1, "isfirsttensorparallelrank": 1, "isgreedysampl": 0, "ishufflelay": 77, "iskvcacheen": 1, "isl": [0, 19, 20, 21, 22, 24, 68, 69, 75], "islastpipelineparallelrank": 1, "isleg": 0, "islicelay": 77, "isload": 1, "islookahead": 0, "islookaheaddecod": 1, "ismedusa": [0, 1], "ismpist": 0, "ismultimod": 1, "isn": 87, "isnon": 1, "isoftmaxlay": 77, "isorchestr": 0, "ispagedkvcach": 1, "isparticip": [0, 88], "ispipelineparallel": 1, "ispoint": 1, "isrnnbas": 1, "issequencefin": [0, 3], "issocketst": 0, "issu": [5, 14, 17, 53, 59, 60, 62, 64, 68, 69, 70, 77, 87], "istensorparallel": 1, "isthreadsaf": 0, "istopk": 0, "istopkandtopp": 0, "istopkortopp": 0, "istopp": 0, "istransformerbas": 1, "istream": [0, 1], "isunsign": 1, "isusebantoken": 0, "isusebanword": 0, "isuseexpliciteosstop": 0, "isusefrequencypenalti": 0, "isusemaxlengthstop": 0, "isuseminlength": 0, "isuseminp": 0, "isusenorepeatngrams": 0, "isuseoccurrencepenalti": 0, "isusepenalti": 0, "isusepresencepenalti": 0, "isuserepetitionpenalti": 0, "isusestopcriteria": 0, "isusestopword": 0, "isusetemperatur": 0, "isusevariablebeamwidthsearch": 0, "iswhisp": 1, "ite": 82, "item": [0, 3, 82], "itensor": [0, 77], "itensorbind": 1, "itensorptr": 1, "iter": [0, 1, 3, 5, 10, 15, 24, 26, 65, 66, 68, 70, 74, 75, 76, 82, 88], "iter_stats_max_iter": 65, "iterationresult": 65, "iterationstat": 0, "iterationtyp": 0, "iterlatencym": [0, 26], "iterlatencymillisec": 88, "iterstat": 0, "iterstatsmaxiter": 0, "iterstatsvec": 0, "ith": 77, "itl": [72, 75, 88], "its": [0, 1, 3, 5, 6, 7, 11, 13, 14, 15, 17, 19, 21, 24, 40, 64, 66, 68, 71, 73, 74, 75, 77, 84, 91, 92, 94], "itself": [3, 82], "itsuji": 68, "iunarylay": 77, "j": [5, 6, 20, 23, 50, 51, 52, 64, 68, 77, 85, 86, 88], "jacobi": 10, "jai": 88, "jamesthez": 88, "jane": 54, "janpetrov": 88, "japanes": [9, 68], "jax": [13, 17], "ji": 77, "jit": [18, 62, 88], "jj": 77, "jk": 77, "jl749": 88, "job": [14, 51, 52], "joint_attention_kwarg": 79, "joint_attn_forward": 78, "journei": [24, 66], "jpg": 68, "json": [0, 1, 3, 13, 26, 29, 30, 31, 34, 35, 40, 47, 65, 67, 68, 83, 88], "json_object": 65, "jsonconfigstr": 0, "jsonl": 68, "jsonseri": 0, "just": [0, 1, 10, 50, 51, 52, 53, 62, 68, 70, 76, 82, 84], "justic": [38, 41, 42, 44, 45, 53], "k": [1, 5, 6, 9, 10, 16, 24, 65, 77, 85, 87, 88, 90, 92], "k_b_proj_tran": 77, "k_dim": 77, "k_proj": [15, 90], "kattent": 1, "kattn_dens": 1, "kattn_k": 1, "kattn_q": 1, "kattn_qkv": 1, "kattn_v": 1, "kauto": 0, "kbatchedpostprocessornam": [0, 3], "kbeamsearch": 0, "kbf16": 0, "kbool": [0, 1], "kbyte_typ": 1, "kc_cache_retention_config": 88, "kcancel": 0, "kchatglm": 1, "kcontext": 1, "kcontext_in_progress": 0, "kcontinu": 1, "kcpu": [0, 1], "kcpu_pin": 0, "kcpu_pinnedpool": 0, "kcross_attn_dens": 1, "kcross_attn_k": 1, "kcross_attn_q": 1, "kcross_attn_qkv": 1, "kcross_attn_v": 1, "kdatatyp": 1, "kdecoder_onli": [0, 11], "kdefault": 0, "kdefault_num_tokens_per_block": 1, "kdefaultbatchsizet": 0, "kdefaultdynamicbatchmovingaveragewindow": 0, "kdefaultgpuspernod": 1, "kdefaultiterstatsmaxiter": 0, "kdefaultlookaheaddecodingngram": 0, "kdefaultlookaheaddecodingverificationset": 0, "kdefaultlookaheaddecodingwindow": 0, "kdefaultmaxadapters": 0, "kdefaultmaxpagesperblockdevic": 0, "kdefaultmaxpagesperblockhost": 0, "kdefaultmaxseqidlemicrosecond": 0, "kdefaultoptimaladapters": 0, "kdefaultprior": 0, "kdefaultrequeststatsmaxiter": 0, "kdefaultretentionprior": 0, "kdisabl": 1, "kdrafttokensextern": 1, "kdynamicpostprocessornameprefix": 0, "keagl": [0, 1], "kebnf_grammar": [0, 3], "keep": [0, 5, 6, 17, 24, 65, 69, 76, 77, 88], "keepdim": 77, "kei": [0, 2, 3, 8, 14, 19, 23, 59, 68, 69, 74, 79, 82, 87, 91, 92, 93], "kenabl": 1, "kencdec": 1, "kencoder_decod": 0, "kencoder_in_progress": 0, "kencoder_onli": 0, "kend_id": 0, "kept": [5, 17, 77], "kequal_progress": 0, "kera": 17, "kernel": [1, 5, 8, 14, 19, 25, 47, 62, 66, 67, 72, 75, 77, 82, 83, 84, 87, 88], "kernel_s": [77, 78], "kexplicitdrafttoken": [0, 1], "kexternaldrafttoken": 0, "key_length": [77, 78], "keyvaluecacheparam": [78, 79], "keyword": [15, 65, 77, 84], "kfirst_come_first_serv": 0, "kfloat": [1, 14], "kfp16": 0, "kfp32": [0, 65], "kfp8": 0, "kgener": 1, "kgeneration_complet": 0, "kgeneration_in_progress": 0, "kglm": 1, "kgpt": 1, "kgpu": [0, 1], "kguaranteed_no_evict": 0, "khalf": 1, "kind": [4, 5, 7, 24, 94], "kinflight": 0, "king": 46, "kint32": [0, 1], "kint64": [0, 1], "kint8": [0, 1], "kinvalid": 1, "kispoint": 1, "kisunsign": 1, "kj": 77, "kjson": [0, 3], "kjson_schema": [0, 3], "kleader": [0, 2], "klength": 0, "klinear": 1, "klookahead": 0, "klookaheaddecod": 1, "kmamba": 1, "kmax_util": 0, "kmaxretentionprior": 0, "kmedusa": [0, 1], "kminretentionprior": 0, "kmla": 0, "kmlp_4h_to_h": 1, "kmlp_gate": 1, "kmlp_gate_up": 1, "kmlp_h_to_4h": 1, "kmlp_router": 1, "kmoe_4h_to_h": 1, "kmoe_gat": 1, "kmoe_h_to_4h": 1, "kmoe_rout": 1, "kmpi": 0, "knegativeinfin": 1, "knob": [0, 65, 76, 77], "knone": 1, "knoop": 1, "knot_finish": 0, "know": [6, 67, 76, 77], "knowledg": 59, "known": [5, 10, 14, 59, 62, 77, 86], "knumflag": 0, "kopt_profiles_split_point": 1, "korchestr": [0, 2], "kosmo": [86, 88], "kpage": 1, "kpin": 1, "kpinnedpool": 1, "kqueu": 0, "krecurr": 1, "krecurrentgemma": 1, "kregex": [0, 3], "kstatic": 0, "kstatic_batch": 0, "kstop_word": 0, "kstructural_tag": 0, "ktimed_out": 0, "ktopk": 0, "ktopktopp": 0, "ktopp": 0, "ktrtpointertyp": 1, "kuint8": [0, 1], "kunderlyingtyp": 1, "kunish": 9, "kunknown": 0, "kunsign": 1, "kusebantoken": 0, "kusebanword": 0, "kuseexpliciteosstop": 0, "kusefrequencypenalti": 0, "kusemaxlengthstop": 0, "kuseminlength": 0, "kuseminp": 0, "kusenorepeatngrams": 0, "kuseoccurrencepenalti": 0, "kusepenalti": 0, "kusepresencepenalti": 0, "kuserepetitionpenalti": 0, "kusestandardstopcriteria": 0, "kusestopword": 0, "kusetemperatur": 0, "kusevariablebeamwidthsearch": 0, "kuvm": [0, 1], "kv": [0, 1, 2, 3, 9, 14, 17, 19, 23, 25, 26, 32, 36, 37, 39, 49, 59, 63, 65, 66, 68, 69, 70, 74, 77, 82, 83, 88, 89, 90, 91, 92, 94], "kv_b_proj": 77, "kv_cach": 0, "kv_cache_block_offset": [77, 78, 82, 87], "kv_cache_block_point": 87, "kv_cache_config": [26, 32, 39, 44, 46, 48, 49, 65, 76, 93], "kv_cache_dtyp": [46, 68, 72, 81, 93], "kv_cache_enable_block_reus": [82, 88], "kv_cache_free_gpu_mem_fract": [18, 69, 76], "kv_cache_free_gpu_memory_fract": [26, 33, 82, 88], "kv_cache_host_memory_byt": 8, "kv_cache_manag": [0, 88, 91, 92, 93, 94], "kv_cache_param": [78, 79, 92], "kv_cache_quant_algo": [13, 54, 65, 68, 72], "kv_cache_quant_mod": [5, 77], "kv_cache_retention_config": 65, "kv_cache_scaling_factor": [5, 13], "kv_cache_typ": [14, 25, 65, 82, 88], "kv_dtype": 79, "kv_event": 46, "kv_head": 78, "kv_host_cache_byt": 8, "kv_lora_rank": [77, 78], "kv_orig_quant_scal": 77, "kv_quant_orig_scal": 77, "kvalue_status_load": 1, "kvalue_status_miss": 1, "kvalue_status_process": 1, "kvcach": [0, 24, 39, 49, 88], "kvcacheconfig": [0, 5, 8, 32, 39, 44, 46, 48, 49, 65, 76, 84], "kvcachecreateddata": [0, 65], "kvcacheev": 0, "kvcacheeventdata": 0, "kvcacheeventdiff": 0, "kvcacheeventmanag": 0, "kvcachehitr": 0, "kvcachehitrateperrequest": 0, "kvcacheindex": 1, "kvcachemanag": [0, 5, 8, 82, 92, 93], "kvcachemetr": 0, "kvcacheparam": 92, "kvcacheremoveddata": [0, 65], "kvcacheretentionconfig": [0, 65], "kvcaches": 0, "kvcachestat": [0, 26], "kvcachestoredblockdata": 0, "kvcachestoreddata": [0, 65], "kvcachetransferend": 0, "kvcachetransferm": 0, "kvcachetransferstart": 0, "kvcachetyp": [1, 65, 82], "kvcachetypefromstr": 1, "kvcacheupdateddata": [0, 65], "kvfactor": 0, "kvheadnum": 77, "kwarg": [15, 17, 65, 77, 78, 79, 82, 88, 90], "kxgrammar": 0, "l": [10, 50, 51, 52, 68, 86], "l2": 25, "l20": 25, "l304": 24, "l345": 24, "l4": 25, "l40": 25, "l440": 24, "l506": 24, "l546": 24, "l823": 24, "lab": 68, "label": [7, 77, 78, 79], "labelembed": 78, "lack": 0, "lambda": [0, 3], "lamportinitializeal": 1, "languag": [0, 6, 10, 14, 16, 19, 24, 66, 67, 77, 85, 86, 88, 91], "language_adapt": [82, 88], "language_adapter_config": 82, "language_adapter_rout": [79, 82], "language_adapter_uid": 82, "language_model": 15, "languageadapterconfig": 82, "languageadapteruid": 0, "larg": [5, 8, 10, 14, 16, 17, 18, 19, 23, 24, 25, 30, 47, 56, 66, 67, 68, 71, 72, 74, 75, 77, 84, 86, 87, 88, 91], "larger": [0, 2, 5, 6, 8, 10, 11, 18, 20, 21, 23, 49, 65, 68, 69, 77, 82, 84, 88], "largest": [6, 19, 20, 21, 77], "last": [0, 1, 3, 5, 9, 10, 12, 24, 65, 74, 76, 77, 79], "last_lay": 82, "last_process_for_ub": 77, "last_token_id": [77, 79, 87], "last_token_ids_for_logit": 79, "last_tokens_id": 77, "lastdraftindic": 1, "lastdraftlen": 1, "lastdraftpath": 1, "lastdrafttoken": 1, "lastgenerationlength": 1, "lastit": 0, "lastpositionidsbas": 1, "lasttokentim": 0, "late": 53, "latenc": [0, 5, 8, 10, 20, 21, 23, 25, 59, 69, 74, 75, 76, 77, 88], "latent": [78, 79], "later": [0, 1, 6, 9, 10, 14, 17, 21, 41, 44, 64, 72, 75, 82, 84, 87, 89], "latest": [0, 27, 60, 83, 88], "latter": [3, 23, 88], "launch": [2, 8, 14, 26, 47, 50, 51, 52, 59, 62, 64, 70, 87, 88, 89], "launch_llama_3": 14, "layer": [0, 1, 2, 4, 5, 6, 7, 9, 10, 12, 13, 14, 15, 25, 65, 71, 77, 82, 83, 84, 85, 87, 88, 90, 92, 93], "layer1": 9, "layer_idx": [9, 12, 77, 82, 90, 92], "layer_names_onli": [25, 65], "layer_norm": [77, 78], "layer_quant_mod": 65, "layer_typ": 82, "layerid": [1, 9], "layeridx": 1, "layernorm": [12, 25, 75, 77, 78, 88], "layernorm_shar": 78, "layernorm_typ": 78, "layernormpositiontyp": 77, "layernormtyp": [77, 78], "layertyp": [1, 7], "layout": [74, 88], "lead": [7, 8, 10, 14, 25, 53, 60, 68, 69, 70, 72, 74, 75], "leader": [0, 82], "learn": [20, 21, 23, 38, 41, 42, 44, 45, 47, 72, 77, 83], "learned_absolut": [13, 77, 78, 79], "least": [3, 5, 17, 18, 26, 53, 74, 82], "leav": [54, 74, 75, 76], "left": [65, 69, 74, 76, 77], "legaci": [15, 76, 80, 88], "len": [1, 68, 77, 82, 94], "length": [0, 1, 5, 8, 18, 19, 20, 21, 22, 23, 24, 25, 26, 48, 65, 68, 69, 70, 73, 75, 76, 77, 82, 84, 87, 88, 92, 93], "length_penalti": [6, 65, 82], "lengthlengthpenalti": 6, "lengthpenalti": [0, 1, 6], "less": [0, 3, 5, 6, 14, 20, 65, 69, 77], "let": [7, 12, 13, 15, 24, 27, 32, 66, 68, 74, 77], "letter": 77, "level": [0, 1, 3, 5, 9, 12, 13, 15, 17, 25, 26, 44, 64, 67, 68, 84, 88, 90], "leverag": [10, 19, 24, 72, 83], "lf": [9, 18, 60, 64], "lfz941": 88, "lh": 1, "lib": [17, 62, 68], "libnvinfer_plugin_tensorrt_llm": 60, "libopenmpi": [61, 62], "librari": [14, 16, 60, 64, 66, 87, 88, 92], "libtensorrt_llm": 60, "licens": [64, 83], "life": 53, "lightweight": 5, "like": [0, 3, 5, 6, 7, 8, 10, 13, 14, 16, 17, 23, 24, 25, 32, 38, 40, 41, 42, 43, 44, 45, 46, 47, 49, 50, 51, 52, 53, 54, 65, 66, 68, 70, 71, 72, 74, 75, 76, 77, 83, 84, 85, 87, 88, 89, 90, 91, 93], "likelihood": [4, 8, 10], "limit": [0, 2, 3, 5, 6, 7, 14, 17, 18, 23, 24, 32, 62, 64, 65, 66, 70, 74, 76, 77, 80, 82, 84, 86, 92], "lin": 19, "line": [8, 18, 23, 68, 70, 72, 75, 84, 88, 93, 94], "linear": [1, 9, 10, 12, 13, 14, 77, 84, 85, 88, 90, 92], "linearactiv": 78, "linearapproximategelu": 78, "linearbas": 78, "lineargeglu": 78, "lineargelu": 78, "linearli": 84, "linearswiglu": 78, "link": [8, 18, 24, 27, 28, 88], "linspac": 77, "linux": [59, 86, 88], "linux_x86_64": 60, "list": [0, 1, 3, 5, 6, 7, 13, 14, 15, 16, 32, 47, 60, 63, 65, 66, 68, 69, 70, 77, 78, 79, 82, 86, 87, 88, 92, 93, 94], "list_siz": 78, "liter": 65, "littl": 75, "live": 84, "livecodebench": 24, "lkm2835": 88, "ll": [23, 26], "llama": [6, 9, 10, 11, 15, 17, 20, 21, 23, 25, 41, 49, 64, 70, 71, 73, 74, 76, 83, 85, 86, 88, 89, 90], "llama2": [5, 9, 19, 20, 88], "llama3": 77, "llama4forconditionalgener": 86, "llama_13b": 21, "llama_70b": 21, "llama_7b": [9, 11], "llama_7b_with_lora_qkv": 9, "llama_model_path": 32, "llamaconfig": [79, 90], "llamaforcausallm": [15, 17, 79, 86], "llamamodel": 79, "llava": [15, 85, 86, 88], "llava_dict": 15, "llavallamamodel": 86, "llavanextforconditionalgener": 86, "llavanextvisionconfig": 79, "llavanextvisionwrapp": 79, "llm": [0, 2, 3, 5, 6, 7, 8, 9, 12, 14, 19, 22, 24, 25, 26, 29, 30, 31, 33, 34, 35, 39, 40, 41, 42, 43, 44, 46, 47, 48, 49, 53, 54, 55, 56, 57, 61, 62, 63, 65, 67, 69, 71, 72, 73, 75, 76, 77, 79, 81, 82, 85, 87, 89, 90, 91, 92, 93, 94], "llm_arg": [65, 69], "llm_engine_dir": 82, "llm_inference_distribut": 64, "llm_kwarg": [39, 49], "llm_mgmn_": 88, "llm_option": 69, "llm_ptq": 89, "llmapi": [3, 26, 32, 39, 40, 44, 46, 48, 49, 50, 51, 52, 54, 65, 69, 72, 88], "llmarg": [65, 69, 88], "llmrequest": [1, 93, 94], "llmrequestptr": 1, "llmrequestst": 94, "lm": 10, "lm_head": [12, 15, 49, 68, 88], "lmm": [6, 68], "lmsy": [39, 49], "ln_emb": 15, "ln_f": [12, 15], "load": [0, 1, 9, 12, 13, 14, 17, 22, 24, 25, 41, 44, 49, 62, 64, 65, 68, 69, 70, 75, 76, 79, 81, 82, 83, 84, 88], "load_format": 65, "load_model_on_cpu": 79, "load_tensor": 15, "load_test_audio": 82, "load_test_data": 82, "load_weight": 90, "loaded_weight": 78, "loader": 88, "loadinprogress": 1, "loadweight": 1, "local": [13, 14, 18, 24, 25, 41, 42, 43, 44, 45, 50, 51, 52, 54, 60, 62, 65, 68, 69, 72, 88, 93], "local_in_featur": 78, "local_layer_idx": 78, "local_model": [50, 51, 52], "local_out_featur": 78, "local_us": [18, 60, 83], "localhost": [26, 29, 30, 31, 33, 34, 35, 55, 56, 57, 83], "localinadapters": 1, "localindim": 1, "localinouts": 1, "localins": 1, "localoutadapters": 1, "localoutdim": 1, "localouts": 1, "localreduct": 24, "localscaless": 1, "localtotals": 1, "locat": [6, 7, 14, 54, 60, 68, 69, 77, 83, 87, 92], "locate_accepted_draft_token": 82, "lock": [62, 68], "lockstep": 0, "log": [0, 1, 5, 25, 26, 27, 50, 51, 52, 54, 65, 68, 77, 83, 84, 88], "log_level": [25, 26], "log_softmax": 77, "logic": [3, 15, 17, 47, 78, 88, 90, 91, 94], "login": [27, 83], "logit": [0, 1, 6, 10, 24, 36, 37, 65, 68, 77, 79, 82, 87, 88], "logits_dtyp": [13, 25, 79], "logits_processor": [47, 65, 82], "logits_processor_map": 82, "logits_processor_nam": 82, "logitspostprocessor": 0, "logitspostprocessorbatch": [0, 3], "logitspostprocessorconfig": [0, 3, 88], "logitspostprocessormap": 0, "logitspostprocessornam": 0, "logitsprocessor": [47, 65, 82, 88], "logitsprocessorlist": 82, "logitsvec": 1, "logn": [77, 88], "logn_scal": 77, "logprob": [0, 1, 32, 48, 65, 83], "logprobs_diff": 65, "logprobscba": 1, "logprobstil": 1, "london": 87, "long": [5, 23, 25, 67, 68, 70, 71, 72, 74, 75, 84, 88], "long_mscal": [77, 78], "long_rop": 77, "long_rope_embed_posit": 78, "long_rope_embed_positions_for_gpt_attent": 78, "long_rope_rotary_cos_sin": 77, "long_rope_rotary_inv_freq": [77, 78], "longer": [0, 6, 8, 24, 65, 69, 74, 77, 94], "longest": [2, 74, 77], "longrop": 77, "longtensor": [82, 90], "look": [0, 3, 17, 22, 60, 66, 68, 88], "lookahead": [0, 1, 36, 37, 59, 65, 88], "lookahead_config": [48, 65, 82], "lookahead_decod": [25, 79], "lookaheadconfig": 0, "lookaheaddecod": 1, "lookaheaddecodingbuff": 1, "lookaheaddecodingconfig": [0, 1, 48, 65], "lookaheadinput": 1, "lookaheadoutput": 1, "lookaheadruntimebuff": 1, "lookaheadruntimeconfig": 1, "lookup": [59, 77, 78, 88], "lookup_plugin": 77, "loop": [0, 3, 6, 14, 15, 76], "lopuhin": 88, "lora": [0, 1, 3, 36, 37, 59, 63, 65, 77, 78, 79, 82, 88], "lora_ckpt_sourc": [25, 82], "lora_config": [53, 65, 79], "lora_dir": [9, 25, 53, 82], "lora_dir1": 53, "lora_dir2": 53, "lora_dir3": 53, "lora_hidden_st": 78, "lora_layer_param": 78, "lora_manag": [53, 65, 82, 88], "lora_param": 79, "lora_plugin": [9, 25, 77, 82], "lora_rank": [9, 77], "lora_request": [53, 65], "lora_runtime_param": 78, "lora_target_modul": [9, 25, 79, 82], "lora_task_uid": 82, "lora_uid": 82, "lora_weights_point": 77, "loracachefullexcept": 1, "loracachepagemanag": 1, "loraconfig": [0, 9, 53, 65, 79, 88], "loraexpectedexcept": 1, "loraid": 0, "loramanag": 82, "loramodulenam": 1, "loraparam": 79, "loraprefetchdir": 0, "lorarequest": [53, 65], "loraruntimeparam": 78, "lorataskidtyp": [0, 1], "loraweight": 9, "loss": [23, 72], "lot": [5, 8, 14, 16], "loudspeak": 21, "lovelac": [66, 86, 88], "low": [5, 12, 17, 18, 23, 24, 25, 59, 77, 88], "low_latency_gemm": 77, "low_latency_gemm_plugin": [25, 68, 72, 78], "low_latency_gemm_swiglu": 77, "low_latency_gemm_swiglu_plugin": [25, 72, 80], "low_rank": 77, "lower": [0, 1, 2, 6, 7, 8, 9, 22, 23, 44, 65, 69, 72, 77, 84], "lru": [1, 8, 77], "lt": 77, "luotuo": 9, "m": [0, 18, 20, 24, 26, 34, 35, 40, 53, 68, 69, 70, 72, 74, 75, 77, 84, 85], "macceptancethreshold": 0, "machin": [8, 18, 23, 47, 88], "madditionalmodeloutput": 0, "made": [47, 66, 88, 94], "mahmoudashraf97": 88, "mai": [0, 1, 2, 3, 5, 8, 9, 10, 13, 14, 15, 17, 18, 24, 25, 27, 50, 51, 52, 60, 62, 64, 67, 68, 69, 70, 75, 76, 77, 78, 80, 84, 87, 88, 90, 91, 92, 93], "main": [3, 6, 19, 22, 24, 30, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 53, 54, 56, 61, 62, 64, 65, 67, 70, 72, 75, 76, 77, 83, 84, 87, 89, 90], "maintain": [2, 9, 19, 20, 23, 68, 72, 85], "major": [17, 24, 54, 66, 69, 84], "make": [1, 2, 5, 7, 9, 14, 17, 18, 23, 24, 27, 28, 48, 53, 59, 60, 66, 68, 70, 76, 77, 83, 87, 88], "make_causal_mask": 78, "makeshap": 1, "mallotedtim": 0, "mallreducecommptr": 1, "mamba": [25, 64, 77, 85, 86, 88], "mamba1": 77, "mamba2": [77, 88], "mamba_conv1d": 77, "mamba_conv1d_plugin": [25, 82], "mamba_vers": 77, "mambaconfig": 79, "mambaforcausallm": 79, "manag": [0, 1, 2, 5, 10, 14, 25, 32, 62, 64, 70, 76, 80, 82, 83, 84, 88, 89, 91, 92], "managedweight": 0, "managedweightsmap": 1, "manageweightstyp": 1, "manageweighttyp": 1, "mandatori": [1, 3, 13], "mani": [0, 5, 8, 10, 14, 17, 25, 28, 54, 65, 69, 72, 74, 76, 77, 86, 87], "manipul": 7, "manner": 7, "mantissa": 20, "manual": [32, 65, 82, 87], "manufactur": 68, "map": [0, 1, 2, 3, 5, 7, 12, 13, 14, 15, 17, 24, 69, 77, 78, 79, 82, 83, 93], "marcellu": 46, "mard1no": 88, "margin": [68, 74], "mark": [1, 7, 74, 77, 87], "mark_as_remov": 7, "mark_output": [3, 77], "markalldon": 1, "markdon": 1, "marks101": 88, "marktaskdon": 1, "mask": [0, 1, 5, 10, 24, 47, 77, 78, 79, 82, 92], "mask_typ": 77, "masked_scatt": 77, "masked_scatter_": 77, "masked_select": [77, 88], "massiv": 18, "master": [71, 72, 73], "mat2": 77, "match": [0, 4, 7, 10, 59, 65, 68, 77, 78, 82, 83, 87, 88], "match_and_rewrit": 7, "materi": 3, "math": [24, 86], "matichon": 88, "matmul": [5, 14, 25, 72, 77, 85], "matric": 4, "matrix": [5, 14, 22, 59, 66, 68, 71, 77, 83, 92], "mattentionconfig": 0, "mattentiontyp": 0, "matter": 8, "matur": 26, "max": [0, 1, 9, 19, 20, 21, 59, 65, 70, 72, 73, 75, 77, 82, 84, 87, 92], "max_all_reduce_block": 1, "max_attention_window": [65, 76, 88], "max_attention_window_s": [5, 76, 77, 82], "max_attn_valu": 78, "max_batch_s": [5, 9, 11, 13, 14, 17, 18, 25, 26, 32, 33, 39, 44, 48, 49, 65, 68, 72, 74, 75, 77, 79, 82, 84, 87, 88, 93], "max_beam_width": [3, 5, 25, 26, 32, 44, 65, 77, 79, 82, 84], "max_block": [77, 94], "max_blocks_per_seq": 82, "max_blocks_per_sequ": 77, "max_boost_slid": 68, "max_cache_storage_gb": 65, "max_context_length": [77, 78, 82, 84], "max_decoder_input_len": 79, "max_decoder_seq_len": 25, "max_dist": [5, 77, 78], "max_draft_len": [25, 39, 49, 65, 79, 81], "max_draft_token": [79, 82], "max_encoder_input_len": [25, 65, 79], "max_gen_token": 79, "max_input_len": [9, 11, 13, 14, 25, 65, 68, 79, 82, 84], "max_input_length": [77, 78, 79, 82], "max_kv_seqlen": 77, "max_lora_rank": [9, 25, 53], "max_low_rank": 77, "max_medusa_token": 82, "max_multimodal_len": 25, "max_new_token": [82, 84], "max_ngram_s": [48, 65], "max_non_leaves_per_lay": [39, 65], "max_num_request": [92, 93, 94], "max_num_token": [18, 25, 26, 32, 33, 44, 65, 68, 72, 74, 75, 79, 84, 88, 92], "max_output_len": [14, 82, 83, 87, 88], "max_period": 78, "max_position_embed": [13, 77, 78, 79], "max_position_embedding_len": 77, "max_power_limit": 68, "max_prompt_adapter_token": 65, "max_prompt_embedding_table_s": [25, 65, 82, 88], "max_record": 65, "max_seq_len": [9, 11, 13, 14, 25, 26, 39, 49, 65, 68, 76, 77, 78, 79, 82, 84, 88, 93], "max_seqlen": [5, 77], "max_seqlen_for_logn_sc": 78, "max_sequence_length": [5, 82], "max_token": [26, 29, 30, 31, 40, 46, 55, 56, 57, 65, 76, 83, 89], "max_tokens_in_paged_kv_cach": [76, 82, 88], "max_util": [0, 65, 76], "max_verification_set_s": [48, 65], "max_window_s": [48, 65], "maxaccepteddrafttokensperstep": 1, "maxacceptedtoken": 1, "maxadapters": 0, "maxattentionwindow": 1, "maxattentionwindowvec": [0, 1], "maxbadwordslen": 1, "maxbatchs": [0, 1, 6], "maxbatchsizeruntim": 0, "maxbatchsizeruntimeupperbound": 0, "maxbatchsizestat": 0, "maxbatchsizetunerrecommend": 0, "maxbeamwidth": [0, 1, 3, 88], "maxdecoderstep": 1, "maxdecodingdrafttoken": 1, "maxdecodingtoken": [0, 1], "maxdraftpathlen": [0, 1], "maxdrafttoken": [0, 1], "maxencoderlen": 1, "maxgenerationlength": 1, "maxgenlengthdevic": 1, "maxgenlengthhost": 1, "maxgentoken": 1, "maxim": [0, 19, 21, 24, 68, 76], "maximum": [0, 1, 2, 3, 5, 6, 18, 21, 25, 26, 65, 68, 69, 72, 77, 78, 82, 84, 87, 88, 93], "maxinputlen": [1, 6], "maxinputlength": 1, "maxlength": 1, "maxlengthstop": 0, "maxlorarank": 1, "maxmedusahead": 1, "maxnewtoken": [1, 88], "maxnonleafnodesperlay": 1, "maxnumactiverequest": 0, "maxnumblock": 0, "maxnumpath": 1, "maxnumsequ": [1, 88], "maxnumtoken": [0, 1], "maxnumtokensruntim": 0, "maxnumtokensstat": 0, "maxnumtokenstunerrecommend": 0, "maxoutputlength": 3, "maxpagesperblock": 1, "maxpagesperblockdevic": 0, "maxpagesperblockhost": 0, "maxpathdraftlen": 1, "maxpathlen": [0, 1], "maxpositionembed": [0, 1], "maxpromptembeddingtables": 1, "maxqueues": 0, "maxseqidlemicrosecond": 0, "maxseqlen": 1, "maxsequencelen": [1, 6], "maxsequencelength": 1, "maxstopwordslen": 1, "maxtoken": [0, 84, 88], "maxtokensperenginestep": 1, "maxtokensperstep": 1, "mb": 84, "mbackend": 0, "mbart": [86, 88], "mbatchingtyp": 0, "mbatchsizet": 0, "mbeamsearchbuff": 1, "mbeamsearchdiversityr": 0, "mbeamwidth": 0, "mbeamwidtharrai": 0, "mbp": 40, "mbuffer": 1, "mbuffermanag": 1, "mc_handl": 1, "mc_ptr": 1, "mc_va": 1, "mcachemap": 1, "mcachemutex": 1, "mcachepagemanag": 1, "mcachest": 0, "mcachetransceiverconfig": 0, "mcapacityschedulerpolici": 0, "mcommmod": 0, "mcommptr": 1, "mcommstat": 0, "mcommtyp": 0, "mcomputecontextlogit": 1, "mcomputegenerationlogit": 1, "mconfig": [0, 1], "mcontextchunkingpolici": 0, "mcontextfmha": 1, "mcontextparallel": 1, "mcopyonpartialreus": 0, "mcpu": 1, "mcpudiff": 1, "mcrosskvcachefract": 0, "mcudagraphcaches": 0, "mcudagraphmod": 0, "mcumlogprobstmp": 1, "md": [2, 10, 12, 24, 77, 88, 91], "mdatatyp": [0, 1], "mdebugconfig": 0, "mdebuginputtensor": 0, "mdebugoutputtensor": 0, "mdebugtensornam": 0, "mdebugtensorsmaxiter": 0, "mdecod": 1, "mdecodedurationm": 0, "mdecoderetentionprior": 0, "mdecoderst": 1, "mdecoderstream": 1, "mdecodingconfig": 0, "mdecodinglayerworkspac": 1, "mdecodingmod": [0, 1], "mdefaulteaglechoic": 1, "mdefaultmedusachoic": 1, "mdefaultposteriorthreshold": 1, "mdevic": 1, "mdevicebuffermanag": 1, "mdevicecacheperc": 0, "mdeviceid": [0, 1], "mdogreedysampl": 1, "mdonetask": 1, "mdprank": 0, "mdpsize": 0, "mdrafttoken": 0, "mdynamicbatchconfig": 0, "mdynamicbatchmovingaveragewindow": 0, "mdynamicdecodelay": 1, "mdynamictreemaxtopk": 0, "me": [30, 53, 54, 56, 83], "meaglechoic": 0, "meagleconfig": 0, "mean": [1, 4, 5, 6, 8, 10, 13, 15, 17, 18, 20, 21, 26, 34, 35, 51, 53, 65, 67, 68, 69, 70, 71, 76, 77, 80, 82, 84], "meaning": [1, 72, 75], "meant": 73, "mearlystop": 0, "measur": [0, 19, 21, 22, 23, 59, 68, 70, 88], "mechan": [3, 14, 93, 94], "media": [68, 88], "media_path": 68, "medium": [23, 87, 88], "medusa": [0, 1, 25, 36, 37, 59, 65, 77, 79, 82, 88], "medusa_choic": [10, 49, 65, 68, 82], "medusa_decode_and_verifi": 82, "medusa_hidden_act": 81, "medusa_logit": 82, "medusa_model_dir": 81, "medusa_output_token": 82, "medusa_path": 82, "medusa_position_offset": 82, "medusa_temperatur": [10, 82], "medusa_topk": 82, "medusa_tree_id": 82, "medusachoic": [0, 1], "medusaconfig": 79, "medusacurtokensperstep": 1, "medusadecodingconfig": [49, 65], "medusaforcausallm": 79, "medusainput": 1, "medusalogit": 1, "medusapath": 1, "medusatargettokensperstep": 1, "medusatreeid": 1, "meet": [23, 77], "membeddingt": 0, "member": [0, 1, 6, 7, 11, 14, 54, 77], "memlock": [60, 87], "memori": [0, 1, 2, 4, 5, 6, 9, 14, 15, 17, 19, 20, 22, 23, 24, 25, 26, 32, 47, 59, 65, 68, 69, 70, 74, 75, 77, 82, 87, 88, 92, 93], "memorypoolfre": [1, 84], "memorypoolreserv": [1, 84], "memorypooltrimto": 1, "memorypoolus": 1, "memorytyp": [0, 1], "memorytypestr": 1, "memtyp": 1, "memusagechang": 84, "menableattentiondp": [0, 1], "menablebatchsizetun": 0, "menableblockreus": 0, "menablechunkedcontext": 0, "menablecontextfmhafp32acc": 0, "menablemaxnumtokenstun": 0, "menablepartialreus": 0, "menabletrtoverlap": 0, "mencodedvocab": 0, "mencoderhiddens": 1, "mengineaddr": 1, "menginebuff": 1, "menginepath": 1, "mengines": 1, "mental": 53, "mention": [6, 17, 18, 32, 72], "menu": [27, 28], "merg": [24, 77], "meshgrid": 77, "meshgrid2d": 77, "messag": [24, 26, 29, 30, 55, 56, 62, 69, 77, 83, 84, 88], "met": [0, 1, 3, 10], "meta": [17, 64, 65, 68, 69, 70, 76, 83, 86], "meta_ckpt_dir": 79, "metadata": [90, 92], "metal": [88, 89], "meth": 64, "method": [0, 1, 3, 5, 6, 10, 11, 13, 14, 17, 19, 32, 47, 62, 68, 82, 85, 87, 88, 90, 91, 93, 94], "metric": [0, 65, 67, 68, 69, 70, 72, 74, 75, 88], "mevent": 1, "meventbuffermaxs": 0, "mexecutionconfig": 1, "mextendedruntimeperfknobconfig": 0, "mfastlogit": 0, "mfinishedstep": 1, "mfirstgentoken": 0, "mfreegpumemoryfract": 0, "mfreepageid": 1, "mfrequencypenalti": 0, "mfuntowicz": 88, "mgathergenerationlogit": 0, "mgemmallreducedtyp": 1, "mgmn": [36, 37], "mgpu": 1, "mgpudiff": 1, "mgpuspernod": 1, "mgpuweightsperc": 0, "mgreedysampl": 0, "mguid": 0, "mguideddecodingconfig": 0, "mguidetyp": 0, "mh": 10, "mh1": 10, "mha": [5, 19, 25, 77, 82, 92], "mhiddens": 1, "mhostcaches": 0, "mi": 85, "mib": 84, "micro": [0, 84], "microbatchid": 0, "microbatchschedul": [91, 94], "microsecond": 0, "microsoft": 13, "middl": 67, "might": [0, 3, 14, 17, 18, 23, 25, 60, 64, 66, 68, 70, 71, 75, 82, 84, 87, 88, 93], "migrat": [17, 80, 88], "million": [54, 68], "millisecond": 0, "millisecondstyp": 0, "mimpl": 0, "min": [0, 1, 6, 20, 24, 68, 70, 75, 77, 87], "min_lat": 77, "min_length": [6, 82], "min_p": [0, 6, 65, 82], "min_token": 65, "mind": [23, 76], "mindim": 1, "mindimfirst": 1, "mini": 88, "minim": [24, 74, 83], "minimum": [0, 5, 6, 65, 68, 69, 72, 77, 82, 84], "minitron": [86, 88], "minittozero": 1, "minlength": [1, 6, 88], "minnormedscorescba": 1, "minor": [54, 88], "minp": [0, 1, 6], "minprogresstask": 1, "minputpack": 1, "minputtokenextraid": 0, "mintoken": [0, 88], "mintpsplitdim": 1, "minut": [0, 23, 70], "mip": 0, "mipcmemoryhandl": 1, "mirco": 0, "mish": 78, "mismatch": [17, 62, 87], "misorchestr": 0, "mispagefre": 1, "miss": [0, 7, 18, 68, 88], "missedblock": 0, "missedblocksperrequest": 0, "mission": 24, "mistral": [4, 64, 68, 72, 75, 85, 86, 88], "mistralai": [68, 86], "mistralforcausallm": 86, "misus": 88, "miterstatsmaxiter": 0, "mitig": [17, 24], "mix": [2, 71, 75, 88], "mixed_precis": 65, "mixer": 88, "mixtral": [4, 9, 64, 68, 72, 75, 85, 86, 88], "mixtralforcausallm": 86, "mixtur": [59, 75, 88], "mjointdecodinginput": 1, "mjointdecodingoutput": 1, "mkdir": 27, "mkdtemp": [41, 44], "mkvcacheconfig": 0, "mkvcachetyp": 1, "mkvfactor": 0, "ml": [77, 88], "mla": [24, 77, 88], "mlayertyp": 1, "mlengthpenalti": 0, "mllama": [86, 88], "mllamaconfig": 79, "mllamaforcausallm": 79, "mllamaforconditionalgener": 86, "mlogit": 0, "mlogitsdtyp": 1, "mlogitspostprocessorconfig": 0, "mlookaheaddecodingconfig": 0, "mlookaheaddecodingmaxnumrequest": 0, "mloramodul": 1, "mloraprefetchdir": 0, "mlp": [9, 12, 14, 15, 25, 77, 87, 88, 90], "mlp_4h_to_h": [9, 25], "mlp_bia": 79, "mlp_gate": [9, 25], "mlp_gate_up": [9, 25], "mlp_h_to_4h": [9, 25], "mlp_output": 87, "mlp_router": [9, 25], "mlphiddens": 1, "mlptype": 77, "mm": 88, "mm_data": 68, "mm_embedding_offload": 82, "mma": 77, "mmanag": 1, "mmanagedweightsmap": 1, "mmanageweightstyp": 1, "mmaxadapters": 0, "mmaxattentionwindow": 0, "mmaxattentionwindowvec": 0, "mmaxbatchs": [0, 1], "mmaxbeamwidth": [0, 1], "mmaxdecodingdecodertoken": 1, "mmaxdecodingdrafttoken": 1, "mmaxdecodingenginetoken": 1, "mmaxdraftpathlen": 1, "mmaxencoderlen": 1, "mmaxinputlen": 1, "mmaxlorarank": 1, "mmaxnonleafnodesperlay": 1, "mmaxnumpackedmask": 1, "mmaxnumpath": 1, "mmaxnumtoken": [0, 1], "mmaxpagesperblock": 1, "mmaxpagesperblockdevic": 0, "mmaxpagesperblockhost": 0, "mmaxpositionembed": 1, "mmaxpromptembeddingtables": 1, "mmaxqueues": 0, "mmaxseqidlemicrosecond": 0, "mmaxsequencelen": 1, "mmaxsequencelength": 1, "mmaxtoken": 0, "mmedusachoic": 0, "mmemorytyp": 1, "mmha": [77, 88], "mminp": 0, "mmintoken": 0, "mmlphiddens": 1, "mmlu": [23, 24, 88], "mmlu_llmapi": 88, "mmmu": 68, "mmodelconfig": [0, 1], "mmodelnam": 1, "mmodelvari": 1, "mmoduleidtomodul": 1, "mmropepositiondelta": 0, "mmroperotarycossin": 0, "mmultiblockmod": 0, "mname": 1, "mnbattentionlay": 1, "mnbhead": 1, "mnbkvheadsperlay": 0, "mnblayer": 1, "mnbrnnlayer": 1, "mngramsiz": 0, "mnorepeatngrams": 0, "mnormalizelogprob": 0, "mnumcopystream": [0, 1], "mnumdecodingenginetoken": 1, "mnumdevicemodulelay": 0, "mnumensurework": 0, "mnumhostmodulelay": 0, "mnumkvheadsperattentionlay": 1, "mnumkvheadspercrossattentionlay": 1, "mnumlanguag": 1, "mnumnod": 0, "mnumputwork": 0, "mnumreturnbeam": 0, "mnumreturnsequ": 0, "mnumsm": 1, "mnumtransformerslay": 1, "modal": 85, "mode": [0, 1, 4, 5, 7, 14, 15, 25, 26, 40, 50, 51, 52, 65, 76, 77, 78, 82, 84, 85, 88, 90], "model": [0, 1, 2, 3, 4, 5, 8, 9, 11, 13, 17, 19, 20, 21, 22, 23, 25, 26, 29, 30, 31, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 61, 62, 65, 66, 67, 70, 73, 76, 77, 78, 80, 81, 82, 84, 85, 89, 92, 93, 94], "model_architectur": 65, "model_cl": 78, "model_cls_fil": 25, "model_cls_nam": 25, "model_config": [25, 65, 82, 90], "model_dir": [9, 11, 12, 13, 14, 15, 17, 49, 50, 68, 71, 79, 81, 83, 87], "model_engin": 93, "model_nam": [51, 69, 82], "model_path": [11, 51, 67, 68], "model_qu": 68, "model_weights_load": [15, 88], "modelconfig": [0, 6, 82, 88, 90], "modelengin": [91, 93], "modelidtomodel": 1, "modeling_deepseekv3": 24, "modeling_llama": 90, "modeling_mymodel": 90, "modeling_opt": 90, "modeling_util": [65, 90], "modelnam": 1, "modelopt": [13, 17, 49, 62, 68, 69, 81, 88], "modelopt_cuda_ext": 62, "modelpath": 0, "modelrunn": [13, 82, 88], "modelrunnercpp": [82, 88], "modelrunnermixin": 82, "modeltyp": [0, 11], "modelvari": 1, "modelweightsformat": 15, "modelweightsload": [15, 88], "modern": 82, "modif": [7, 14], "modifi": [3, 7, 60, 68, 72, 75, 76, 87, 88], "modul": [0, 1, 5, 6, 12, 13, 14, 15, 24, 25, 59, 60, 65, 75, 77, 78, 79, 81, 82, 87, 88, 90], "modular": 66, "module1": 24, "module10": 24, "module11": 24, "module12": 24, "module13": 24, "module2": 24, "module3": 24, "module4": 24, "module5": 24, "module6": 24, "module7": 24, "module8": 24, "module9": 24, "module_id": 9, "moduleid": [1, 9], "moduleidtomodel": 1, "modulelist": 90, "moduletyp": 1, "modulo": 77, "moe": [9, 15, 24, 25, 45, 59, 65, 75, 77, 79, 88], "moe_4h_to_h": [9, 25], "moe_allreduce_residual_rms_norm": 77, "moe_backend": 18, "moe_cluster_parallel_s": 65, "moe_ep_s": 4, "moe_expert_parallel_s": [45, 65], "moe_gat": [9, 25], "moe_h_to_4h": [9, 25], "moe_plugin": 25, "moe_rout": [9, 25], "moe_tensor_parallel_s": [45, 65], "moe_tp_siz": 4, "moeconfig": 79, "moetopk": 88, "moment": 3, "monboardblock": 0, "monitor": 25, "monitor_memori": [25, 65], "monolith": 5, "monost": 0, "month": 68, "mopenipc": 1, "moptimaladapters": 0, "morchestratorconfig": 0, "morchleadercomm": 0, "more": [0, 1, 2, 3, 4, 5, 6, 7, 10, 12, 13, 14, 19, 20, 21, 23, 24, 25, 26, 32, 36, 46, 47, 54, 60, 65, 66, 68, 69, 70, 72, 74, 75, 76, 77, 83, 84, 87, 88, 90, 92, 94], "most": [0, 1, 6, 10, 14, 17, 19, 20, 21, 23, 24, 38, 41, 42, 44, 45, 65, 67, 73, 75, 76, 77, 84, 87, 88], "mount": [26, 50, 51, 52], "mount_dest": [50, 51, 52], "mount_dir": [50, 51, 52], "moutdim": 1, "moutdimfirst": 1, "moutputbeamhypothes": 1, "mouttpsplitdim": 1, "move": [0, 1, 17, 47, 65, 66, 77, 87, 88], "movement": 14, "mownsev": 1, "mownsstream": 1, "mp4": [30, 56], "mpageblock": 1, "mpagedcontextfmha": 1, "mpagedst": 1, "mpagemanagerconfig": 1, "mpagesmutex": 1, "mpagewidth": 1, "mparallelconfig": 0, "mparticipantid": 0, "mpeftcacheconfig": 0, "mpi": [0, 1, 2, 6, 14, 16, 17, 25, 26, 50, 51, 52, 62, 65, 67, 68, 70, 77, 87, 88], "mpi4pi": [64, 70, 87, 88], "mpi_abort": 64, "mpi_barri": 17, "mpi_comm_world": [6, 64], "mpi_group_barri": 1, "mpicomm": 0, "mpicommsess": 65, "mpin": 1, "mpinneddiff": 1, "mpinnedpool": 1, "mpinnedpooldiff": 1, "mpipelineparallel": [0, 1], "mpirun": [13, 14, 64, 70, 87, 88], "mpisess": 65, "mpistat": 0, "mpointer": 1, "mpool": 1, "mport": 0, "mposteriorthreshold": 0, "mppreducescatt": 1, "mprecis": 1, "mpresencepenalti": 0, "mprocessorbatch": 0, "mprocessormap": 0, "mprompttableoffload": 0, "mpt": [23, 85, 86, 88], "mptforcausallm": 79, "mptmodel": 79, "mqa": [5, 19, 22, 24, 25, 77, 88, 92], "mquantmod": 1, "mrank": [0, 1], "mrecvpollperiodm": 0, "mrepetitionpenalti": 0, "mreplic": 0, "mreqid": 0, "mrequeststatsmaxiter": 0, "mrnnconfig": 1, "mrope": [0, 77], "mrope_param": [78, 82], "mrope_position_delta": [77, 78, 82], "mrope_rotary_cos_sin": [77, 78], "mrope_rotary_cos_sin_s": 79, "mropeconfig": 0, "mropeparam": [78, 82], "mropepositiondelta": 0, "mroperoratysinco": 0, "mrotaryembeddingdim": 1, "mruntimedefault": 1, "mruntimestream": 1, "msamplingconfig": 1, "mscale": 77, "mscale_all_dim": 77, "mschedulerconfig": 0, "msecondaryofflineminprior": [0, 65], "msecondaryoffloadminprior": 0, "mseed": 0, "mselfidx": 0, "msg": [0, 1, 24], "msinktokenlength": 0, "msizeperhead": [0, 1], "mskipcrossattnblock": 1, "msl": 1, "mslotsperpag": 1, "mspawnprocess": 0, "mspeculativedecodingconfig": 0, "mspeculativedecodingmod": 1, "mspeculativedecodingmodul": 1, "mstate": [0, 1], "mstoptokenid": 0, "mstream": 1, "mt5": 86, "mtag": 0, "mtaskid": 0, "mtemperatur": 0, "mtensor": 0, "mtensorparallel": [0, 1], "mtoken": 0, "mtokenizerstr": 0, "mtokenrangeretentionconfig": 0, "mtokensperblock": [0, 1], "mtopk": 0, "mtopp": 0, "mtoppdecai": 0, "mtoppmin": 0, "mtoppresetid": 0, "mtotalnumpag": 1, "mtp": [18, 65, 88], "mtp3_autoregress": 24, "mtp3_top1": 24, "mtp3_top10": 24, "mtp3_top15": 24, "mtp3_vanilla": 24, "mtpdecodingconfig": 65, "mtprank": 1, "mtrimpool": 1, "mtype": 1, "much": [8, 14, 67, 69, 74, 84], "mul": 77, "multi": [0, 2, 3, 4, 6, 8, 9, 10, 13, 16, 17, 19, 25, 30, 50, 51, 52, 56, 59, 60, 64, 65, 70, 77, 79, 84, 85, 88, 92], "multi_block_mod": [5, 65, 82, 88], "multiblockmod": 0, "multidimension": 77, "multihead": [14, 19], "multimod": [0, 25, 58, 68, 82, 86, 88], "multimodalembed": 0, "multimodalmodelrunn": 82, "multinod": 71, "multinomi": 6, "multipl": [0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 14, 15, 24, 25, 36, 37, 65, 66, 70, 71, 72, 74, 77, 78, 82, 83, 87, 88, 92], "multiple_profil": [25, 68, 72, 75, 88], "multipli": [5, 15, 77], "multiply_and_lora": 78, "multiply_collect": 78, "multiprocessor": 14, "munsign": 1, "musecrossattent": 1, "musedynamictre": 0, "musegemmallreduceplugin": 1, "musegptattentionplugin": 1, "musegpudirectstorag": 0, "museloraplugin": 1, "musemambaconv1dplugin": 1, "musemrop": 1, "musepositionembed": 1, "museshapeinfer": 1, "musetokentypeembed": 1, "must": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 14, 16, 25, 26, 28, 40, 65, 72, 77, 78, 80, 82, 85, 87], "mutabl": [0, 1], "mutablepageptr": 1, "mutex": 1, "mutual": [6, 85], "muvm": 1, "muvmdiff": 1, "mverificationsets": 0, "mversion": 1, "mvocabs": 1, "mvocabsizepad": 1, "mweight": 0, "mwindows": 0, "mworkerexecutablepath": 0, "mworldconfig": 1, "my": [1, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 54, 61, 62, 68, 83, 89], "my_faster_on": 32, "my_model": 12, "my_profile_export": [26, 34, 35], "myattent": 90, "mybatchedlogitsprocessor": 47, "myconfig": 90, "mydecoderlay": [12, 90], "mylogitsprocessor": 47, "mymodel": [12, 90], "mymodelforcausallm": [12, 90], "n": [1, 2, 5, 9, 10, 13, 14, 26, 38, 40, 41, 42, 43, 44, 45, 47, 50, 51, 52, 53, 54, 64, 65, 68, 70, 74, 77, 78, 79, 84, 85, 87, 88], "n_worker": 65, "na": [68, 88], "naiv": 75, "naivepatternrewriter_replaceaddwithsub": 7, "name": [0, 1, 3, 6, 7, 9, 13, 14, 26, 27, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 51, 54, 61, 62, 64, 65, 68, 69, 70, 77, 79, 80, 81, 82, 83, 87, 88, 89, 90], "named_network_output": 87, "named_paramet": 15, "namespac": [0, 1, 64, 79], "nation": 68, "nationwid": 68, "nativ": [17, 20, 88, 90], "native_quant_flow": 79, "natur": [17, 30, 56, 70], "naur": [0, 3, 65], "nb": 79, "nbattentionlay": [0, 1], "nbdim": 1, "nbhead": 1, "nbkvhead": [0, 1], "nbkvheadperlay": 0, "nblayer": 1, "nbrnnlayer": 1, "nccl": [14, 24, 25, 77, 87, 88], "nccl_p2p_level": 88, "nccl_plugin": 25, "ncclplugin": 14, "ncclrecv": 77, "ncclsend": 77, "nd": [68, 77], "ndarrai": [77, 78, 82], "ndim": 77, "nearest": 77, "nearli": [7, 20], "necess": 10, "necessari": [1, 4, 10, 24, 53, 72, 77, 88, 93], "necessarili": [1, 14, 84], "need": [1, 2, 3, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 24, 26, 27, 32, 36, 40, 45, 50, 51, 52, 53, 60, 61, 62, 64, 65, 66, 68, 69, 70, 71, 72, 74, 75, 76, 77, 79, 80, 82, 83, 84, 87, 88, 90, 91, 92, 93, 94], "needed_block": 94, "needsdecoderprologu": 1, "needskvcacherewind": 1, "neg": [1, 76, 77], "neglig": [8, 23, 74], "neither": [3, 77, 84], "nemo": [13, 16, 25, 66, 70, 82, 85, 86, 88], "nemo_ckpt_dir": 79, "nemo_prompt_convert": 82, "nemotron": [86, 88], "nemotron_na": 88, "nemotronforcausallm": 86, "nemotronna": [86, 88], "nemotronnasforcausallm": 86, "neox": [5, 6, 85, 86, 88], "nest": 7, "net": [8, 87], "net_guard": 7, "network": [3, 4, 5, 7, 14, 16, 17, 25, 40, 77, 83, 84, 85, 87, 88], "neural": [4, 7, 14, 83, 88], "neva": [86, 88], "never": [7, 68, 76], "new": [0, 1, 3, 5, 6, 7, 8, 9, 10, 11, 17, 20, 21, 24, 26, 27, 29, 31, 38, 41, 42, 43, 44, 45, 47, 55, 57, 59, 60, 64, 65, 66, 74, 75, 77, 82, 83, 88, 89, 91, 93], "new_decoder_architectur": [13, 79], "new_generated_id": 82, "new_input": 7, "new_out": 7, "new_shap": 77, "new_tensor": 77, "new_token": 82, "new_workflow": 88, "newactiverequestsqueuelatencym": [0, 26], "newer": [86, 88], "newest": 21, "newli": [0, 65, 74], "newsiz": 1, "newtoken": 1, "newtokensstep": 1, "newtokensvec": 1, "newvalu": 0, "next": [1, 9, 10, 14, 17, 20, 59, 60, 66, 71, 72, 74, 75, 76, 82, 84, 86, 88], "next_logit": 82, "next_medusa_input_id": 82, "next_medusa_logit": 82, "next_step_buff": 82, "next_step_tensor": 82, "nextdraftindic": 1, "nextdraftlen": 1, "nextdraftpath": 1, "nextdraftprob": 1, "nextdrafttoken": 1, "nextdrafttokenslen": 1, "nextflattoken": 1, "nextgenerationlength": 1, "nextpositionoffset": 1, "ngc": [61, 62, 83, 88, 89], "ngoanpv": 88, "ngram": [0, 6, 65], "ngramsiz": 0, "ngroup": 77, "nhead": 77, "nhere": 40, "ni": [40, 85], "nine": 83, "nj": 43, "njane": [38, 41, 42, 43, 44, 45, 47], "njason": 53, "nmh": 82, "nmt": [82, 86, 88], "nn": [77, 90], "no_quant": 65, "no_repeat_ngram_s": [6, 65, 82], "no_schedule_after_st": 94, "no_schedule_until_st": 94, "noauxtckernel": 24, "node": [0, 2, 6, 16, 25, 50, 51, 52, 59, 64, 65, 67, 70, 71, 77, 82, 85, 87, 88], "noexcept": [0, 1], "nomin": [38, 41, 42, 43, 44, 45], "non": [0, 2, 5, 11, 14, 17, 23, 24, 25, 47, 77, 87, 88], "non_block": 47, "non_gated_vers": 77, "none": [1, 6, 7, 12, 15, 17, 25, 26, 32, 46, 47, 48, 49, 53, 54, 65, 68, 70, 74, 77, 78, 79, 80, 81, 82, 87, 88, 90, 92], "nonetyp": [65, 82], "nonzero": 77, "nor": 84, "norepeatngrams": [0, 1, 6], "norm": [15, 18, 51, 67, 68, 69, 70, 77, 88, 90], "norm_before_bmm1": [78, 79], "norm_elementwise_affin": 78, "norm_ep": 78, "norm_epsilon": [13, 79], "norm_factor": 5, "norm_num_group": 78, "norm_pre_residual_weight": 77, "norm_quant_fus": 25, "norm_typ": 78, "norm_weight": 77, "normal": [0, 6, 8, 9, 11, 23, 24, 65, 68, 77, 84, 88], "normalize_log_prob": 65, "normalize_weight": 9, "normalized_shap": [77, 78], "normalizelogprob": [0, 1], "normedscorescba": 1, "north": [12, 14, 87], "northeastern": 83, "not_op": 77, "notabl": 23, "note": [1, 2, 7, 8, 9, 10, 14, 18, 21, 23, 24, 25, 28, 32, 46, 50, 51, 52, 54, 59, 60, 68, 69, 72, 74, 76, 77, 80, 82, 84, 85, 86, 87, 89, 90, 93], "notic": [46, 53], "notimplementederror": 17, "nougat": [85, 86, 88], "nour": 54, "now": [6, 10, 13, 15, 19, 24, 66, 68, 74, 80, 83, 84, 88], "np": 77, "npy": 82, "npytorch_backend_config": 26, "nsight": 59, "nsy": 67, "ntask": [14, 26, 50, 51, 52], "null": [1, 13, 68, 83], "nullopt": [0, 1], "nullptr": [0, 1], "num": [0, 1, 18, 49, 51, 59, 65, 67, 68, 69, 70, 72, 73, 75], "num_attention_head": [13, 77, 78, 79], "num_aud_token": 82, "num_beam": [6, 82], "num_beam_group": 6, "num_block": [82, 93], "num_blocks_per_cache_level": 46, "num_bucket": [77, 78], "num_channel": [78, 79], "num_class": 78, "num_context": 92, "num_ctx_token": 92, "num_draft_token": [0, 77, 82], "num_eagle_lay": [39, 65], "num_embed": 78, "num_experts_per_tok": 4, "num_gener": 92, "num_group": [77, 78], "num_head": [5, 15, 77, 82, 92], "num_hidden_lay": [13, 79, 90, 93], "num_imag": 82, "num_img_token": 82, "num_key_value_head": [13, 79, 93], "num_kv_head": [77, 78, 82, 92, 93], "num_kv_heads_origin": 77, "num_kv_heads_per_cross_attn_lay": 82, "num_kv_heads_per_lay": 82, "num_lay": [77, 78, 82, 93], "num_ln_in_parallel_attn": 79, "num_local_block": 78, "num_local_expert": 4, "num_lora_module_lay": 9, "num_lora_modules_lay": 9, "num_medusa_head": [49, 65, 79, 81, 82], "num_medusa_lay": [79, 81], "num_multimodal_token": 0, "num_nextn_predict_lay": [18, 65], "num_orig_po": 77, "num_po": 77, "num_postprocess_work": 26, "num_profil": 79, "num_q_head": 24, "num_request": [18, 68, 69], "num_return_sequ": [82, 88], "num_sampl": 67, "num_task": 78, "num_token": [5, 24, 77, 92], "num_tokens_per_block": [77, 93], "num_tokens_per_task": 78, "num_video": 82, "numactiverequest": 0, "numattentionhead": 1, "numavailablepag": 1, "numbeamscba": 1, "number": [0, 1, 2, 3, 4, 5, 6, 10, 14, 18, 22, 24, 25, 26, 47, 50, 51, 52, 65, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 82, 84, 85, 87, 88, 90, 92, 93], "numblockspercachelevel": 0, "numcompletedrequest": 0, "numcontextrequest": [0, 1], "numcopystream": [0, 1], "numctxsequ": 1, "numctxtoken": 0, "numdevicemodulelay": 0, "numdrafttoken": 1, "numdrafttokenshost": 1, "numeaglelay": 1, "numel": 82, "numensurework": 0, "numer": [6, 24, 59, 68, 83, 86], "numexpert": 1, "numgeneratedtoken": 0, "numgenrequest": 0, "numgensequ": 1, "numgentoken": 0, "numhead": 6, "numhostmodulelay": 0, "numkvattentionhead": 1, "numkvhead": 6, "numlanguag": 1, "numlay": 6, "nummissedblock": 0, "numnewactiverequest": 0, "numnewallocatedblock": 0, "numnewtokenscumsum": 88, "numnod": [0, 88], "numpag": 1, "numpausedrequest": 0, "numpi": [9, 77, 82], "numputwork": 0, "numqueuedrequest": [0, 88], "numreturnbeam": 0, "numreturnsequ": [0, 1, 3], "numreusedblock": 0, "numscheduledrequest": 0, "numsequ": 1, "numslot": 1, "numtoken": 1, "numtotalallocatedblock": 0, "numtransformerslay": 1, "nvcc": 18, "nvcr": 88, "nvfp4": [24, 25, 54, 59, 65, 68, 88, 89], "nvidia": [13, 14, 16, 17, 18, 19, 20, 21, 23, 25, 27, 29, 30, 31, 33, 34, 35, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 60, 62, 66, 68, 69, 70, 75, 77, 83, 84, 86, 87, 88, 89], "nvila": [86, 88], "nvinfer1": [0, 1], "nvl": [1, 25, 88], "nvl36": 71, "nvl72": 71, "nvlink": [2, 6, 70, 71, 73, 88], "nvswitch": [14, 24], "nyou": 40, "o": [0, 1, 7, 9, 17, 22, 24, 50, 51, 52, 67, 87], "o_proj": 15, "oai": [30, 56], "obei": 87, "object": [0, 1, 3, 8, 12, 14, 15, 17, 32, 40, 65, 77, 78, 79, 80, 82, 83, 84, 91], "observ": [46, 69], "obtain": [2, 16, 69, 77], "occas": 87, "occasion": 88, "occup": [5, 84], "occupi": [23, 84], "occur": [6, 8, 93, 94], "odd": 47, "off": [8, 67, 72, 74, 75, 84, 88], "offer": [14, 16, 23, 24, 66, 92], "offic": 40, "officenetsecur": 40, "offici": [5, 18, 68], "offlin": [12, 21, 36, 68, 69, 88], "offload": [0, 11, 25, 59, 65, 88], "offset": [1, 77, 82, 85, 88], "offsetdim": 1, "ofitensor": 0, "often": [0, 3, 10, 19, 23, 24, 65, 71, 72, 77], "ok": 87, "okai": 46, "old": [7, 9, 87], "older": [8, 17, 60, 86], "oldest": 9, "oldvalu": 0, "omit": [1, 3, 17, 77], "ompi": [62, 87], "onboard": [0, 8, 65, 84], "onboard_block": 65, "onboardblock": 0, "onc": [0, 3, 5, 6, 7, 14, 16, 60, 64, 65, 72, 77, 84], "one": [0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 13, 14, 15, 17, 19, 24, 25, 26, 27, 53, 64, 65, 68, 70, 71, 72, 75, 76, 77, 78, 80, 82, 84, 87, 88, 90, 94], "ones": [0, 9], "oneshot": [24, 77], "oneshotallreduc": 24, "oneshotar": 24, "onevis": [86, 88], "ongo": [17, 54], "onli": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 17, 18, 23, 25, 26, 32, 47, 54, 59, 64, 65, 68, 69, 70, 71, 72, 74, 75, 76, 77, 78, 80, 82, 84, 86, 88, 91, 94], "onlin": [16, 21, 36], "only_cross_attent": 78, "onnx": [25, 77], "onnx__gathernd": 77, "onto": 6, "oom": [18, 19, 22, 84], "ootb": 88, "op": [1, 7, 77, 88], "op_and": 77, "op_or": 77, "op_xor": 77, "opaqu": 7, "opaque_st": 65, "open": [6, 19, 24, 54, 66, 67, 87, 88], "openai": [26, 58, 83, 88], "openipc": 1, "openmpi": 88, "opensora": 88, "openssh": 27, "oper": [0, 1, 3, 5, 6, 7, 10, 13, 14, 15, 24, 25, 47, 65, 68, 71, 72, 75, 77, 83, 84, 86, 88, 91, 92, 93], "opportun": 68, "opt": [3, 13, 23, 27, 77, 85, 86, 87, 88], "opt_batch_s": [65, 79], "opt_num_token": [25, 65, 79], "optforcausallm": [13, 79], "optim": [1, 2, 3, 6, 7, 10, 14, 16, 17, 19, 20, 21, 22, 23, 25, 41, 47, 49, 60, 64, 66, 68, 69, 71, 72, 73, 77, 83, 84, 86, 87, 88, 89, 91, 92, 93], "optimaladapters": [0, 1], "option": [0, 1, 3, 6, 7, 10, 12, 17, 20, 25, 26, 32, 47, 51, 53, 59, 62, 65, 67, 68, 69, 70, 71, 73, 74, 77, 80, 82, 84, 87, 88, 90, 92, 93], "optionalbufferptr": 1, "optionaltensorptr": 1, "optmodel": 79, "optvec": 1, "orchestr": [0, 2, 10, 87, 88], "orchestratorconfig": 0, "orchleadercomm": 0, "order": [0, 1, 2, 5, 15, 19, 65, 68, 69, 72, 76, 77, 78, 84], "org": [0, 1, 4, 9, 25, 61, 62, 77, 85], "organ": [66, 93], "origin": [5, 7, 9, 77, 88, 90], "original_max_position_embed": [77, 78], "originaltemperatur": 1, "oserror": 88, "osl": [19, 20, 21, 22, 24, 68, 69, 75], "ostream": [0, 1], "other": [0, 1, 2, 3, 4, 5, 6, 8, 10, 14, 15, 17, 19, 24, 25, 32, 44, 46, 50, 51, 52, 54, 60, 64, 65, 66, 69, 70, 71, 72, 74, 75, 76, 77, 80, 84, 87, 88, 92, 94], "other_audio_input": 82, "other_decoder_input": 82, "other_vision_input": 82, "othercach": 1, "otherwis": [0, 1, 3, 5, 6, 32, 65, 68, 77, 82, 87, 92], "our": [18, 23, 24, 38, 40, 41, 42, 44, 45, 68, 69, 72, 74, 75, 77, 86, 87, 88, 90], "out": [0, 1, 2, 9, 17, 19, 20, 21, 22, 24, 36, 50, 51, 52, 64, 67, 69, 72, 74, 75, 77, 83, 84, 88], "out_bia": 78, "out_channel": 78, "out_context_dim": 78, "out_dim": 78, "out_fatur": 13, "out_featur": [13, 14, 78], "out_hidden_s": 77, "out_of_tree_exampl": 90, "out_point": 77, "out_tp": [19, 22], "outdim": 1, "outdimfirst": 1, "outer": 77, "outlin": 67, "output": [0, 1, 2, 5, 6, 7, 8, 9, 10, 14, 18, 19, 20, 21, 22, 23, 24, 25, 26, 32, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 51, 53, 54, 61, 62, 65, 67, 69, 70, 71, 72, 73, 75, 76, 77, 78, 82, 83, 87, 88, 89, 91, 92, 94], "output_cum_log_prob": 82, "output_dim": 78, "output_dir": [9, 11, 12, 13, 14, 17, 25, 68, 71, 79, 81, 83, 87], "output_dtyp": [77, 78], "output_generation_logit": 82, "output_id": 82, "output_log_prob": 82, "output_multiplier_scal": 79, "output_pad": [77, 78], "output_s": 78, "output_seqlen": [19, 22], "output_sequence_length": 82, "output_timing_cach": [25, 65], "output_token": 68, "outputbuff": 1, "outputconfig": [0, 3, 32, 88], "outputidscba": 1, "outputlen": 0, "outputlogprob": 1, "outputtokenid": [0, 3], "outsid": [10, 16, 17, 92], "outsiz": 1, "outtpsplitdim": 1, "outweigh": 71, "over": [0, 1, 8, 10, 15, 18, 20, 21, 23, 24, 28, 67, 68, 71, 74, 75, 77, 88], "overal": [3, 5, 8, 10, 18, 66, 71, 72, 74, 75, 76, 90], "overcom": [5, 14, 24], "overflow": 1, "overhead": [3, 14, 24, 71, 88, 92], "overiew": 68, "overlap": [0, 2, 10, 18, 24, 88, 94], "overload": [0, 1], "overrid": [1, 15, 17, 32, 77, 82], "override_field": 79, "overshadow": 71, "oversubscrib": [64, 70], "overview": [3, 18, 23, 59, 60, 67, 68, 70, 89, 91], "overwhelm": 53, "overwrit": [5, 26], "own": [0, 1, 2, 8, 10, 13, 14, 15, 16, 17, 18, 32, 60, 90], "ownership": 0, "ownsev": 1, "ownsstream": 1, "p": [0, 6, 10, 16, 27, 50, 51, 52, 65, 79, 82, 88], "p2p": 77, "p50": [68, 69], "p90": [68, 69, 70], "p95": [68, 69, 70], "p99": [68, 69, 70], "p_max": 0, "p_x": 0, "pack": [0, 1, 6, 25, 59, 76, 77, 79, 84, 90], "packag": [3, 60, 61, 62, 68, 70, 87, 88], "packed_length": 79, "packedinput": 1, "packedmask": 1, "packedmaskhost": 1, "packedmaskhostcopi": 1, "packedmasksdevic": 1, "packedpositionid": 1, "pad": [0, 1, 6, 7, 9, 25, 26, 59, 65, 66, 77, 78, 82, 84, 88], "pad_id": [65, 82], "pad_lda": 78, "pad_ldc": 78, "pad_token_id": 82, "padding_2d": 77, "padding_back": 77, "padding_bottom": 77, "padding_front": 77, "padding_left": 77, "padding_mod": 78, "padding_right": 77, "padding_top": 77, "padid": 0, "page": [1, 2, 6, 8, 14, 21, 25, 59, 64, 68, 70, 72, 77, 83, 84, 88, 92], "paged_context_fmha": [72, 88], "paged_kv_cach": [9, 25, 68, 82], "paged_st": [25, 82], "pagedcontextfmha": 1, "pagedkvcach": 6, "pagedst": 1, "pageid": 1, "pageidx": 1, "pagemanagerconfig": 1, "pageptr": 1, "pagewidth": 1, "pair": [0, 1, 19, 72, 75, 77], "pale": 46, "paper": [2, 9, 10, 20, 85, 92], "par": [74, 75], "parallel": [0, 2, 3, 5, 6, 10, 13, 14, 18, 19, 21, 22, 26, 36, 37, 45, 47, 59, 65, 69, 72, 73, 77, 78, 79, 84, 88, 90, 94], "parallel_attent": [13, 79], "parallelconfig": [0, 88], "param": [0, 1, 15, 41, 42, 43, 45, 46, 54, 65, 77, 78, 79, 82], "paramet": [0, 1, 3, 4, 5, 8, 9, 10, 11, 13, 14, 15, 17, 18, 25, 26, 50, 65, 68, 71, 72, 73, 76, 77, 78, 79, 82, 84, 88, 92], "parametr": 82, "parent": [0, 1, 15, 17], "parent_hash": 46, "parenthash": 0, "parentid": 1, "pari": [38, 41, 42, 43, 44, 45, 54], "pars": [1, 65], "parse_arg": 49, "parser": [26, 49, 58], "part": [1, 3, 4, 7, 14, 15, 17, 59, 60, 64, 65, 66, 69, 74, 75, 76, 77, 82, 84], "part2": 88, "parti": 88, "partial": [0, 4, 8, 14, 24, 65, 71], "particip": [0, 54, 77, 88], "participantid": [0, 2], "particular": [0, 3, 64, 73, 74, 75, 83], "particularli": [24, 60, 75, 93], "partit": [5, 9, 14, 50, 51, 52], "pass": [0, 1, 3, 5, 7, 8, 9, 10, 14, 15, 32, 47, 53, 54, 65, 67, 68, 70, 72, 74, 75, 77, 78, 79, 82, 84, 88, 89, 90, 91, 92, 94], "past": [0, 5], "past_key_valu": [77, 78], "past_key_value_length": 78, "past_key_values_length": 78, "past_kv_length": 82, "past_sequence_length": 82, "patch": [78, 82], "patch_siz": [78, 79], "path": [0, 1, 3, 5, 10, 13, 15, 18, 25, 26, 32, 41, 42, 43, 44, 45, 49, 50, 51, 52, 54, 60, 64, 65, 67, 68, 69, 70, 72, 77, 82, 88], "path_to_llama_from_hf": 91, "path_to_meta_llama_from_hf": 64, "path_to_trt_engin": 64, "pathlib": [49, 65], "pathlik": 79, "pathorn": 88, "pathsoffset": 1, "pattern": [4, 24, 59, 65, 77, 88], "patternanalyz": 7, "patternrewrit": 7, "paus": [0, 76, 94], "paused_request": 94, "pcie": 25, "pdf": [0, 4, 9], "pdl": [24, 88], "peak": [0, 18, 19, 20, 24, 69], "peft": 65, "peft_cache_config": [32, 44, 65], "peftcacheconfig": [0, 65], "peftcachemanag": [0, 88], "penal": [0, 6, 65], "penalti": 88, "penalty_alpha": 6, "pend": 94, "pending_request": 94, "per": [0, 1, 3, 5, 6, 10, 14, 17, 18, 19, 21, 22, 24, 25, 26, 50, 51, 52, 65, 68, 69, 70, 71, 72, 77, 78, 84, 85, 88], "per_channel": 85, "per_group": 85, "per_token": 85, "per_token_scal": 77, "perceiv": 20, "percent": [0, 11], "percentag": [9, 11, 68, 69, 70], "percentil": [68, 88], "perf": [0, 18, 26, 58, 65, 77, 88], "perf_best_practic": 88, "perform": [0, 1, 2, 3, 5, 6, 7, 9, 14, 15, 16, 17, 19, 21, 22, 25, 26, 32, 60, 64, 65, 66, 68, 69, 71, 74, 76, 77, 82, 83, 86, 88, 90, 92, 93], "performantli": 19, "permut": 77, "persimmon": 88, "persist": [23, 64], "person": [27, 53], "phase": [0, 2, 7, 10, 19, 22, 24, 25, 59, 68, 73, 74, 75, 76, 77, 84, 88, 92, 93], "phi": [64, 77, 85, 86, 88], "phi3config": 79, "phi3forcausallm": 79, "phi3model": 79, "phiconfig": 79, "phiforcausallm": 79, "phimodel": 79, "physic": [77, 84], "picasso": 54, "pick": 74, "pickl": 88, "piec": 74, "pin": [0, 1, 8], "ping": 88, "pinnedmemusag": 0, "pinnedpool": 1, "pip": [18, 26, 60, 61, 62, 83, 88], "pip3": [61, 62], "pipelin": [0, 1, 3, 6, 14, 19, 22, 25, 26, 45, 65, 68, 69, 73, 84, 88, 94], "pipeline_parallel_s": [45, 65, 71, 72], "pipelineparallel": [0, 1, 6], "pipelineparallelismrank": 1, "pitfal": [8, 17], "pixart": 78, "pixartalphatextproject": 78, "pixel_valu": 79, "pl": [62, 68], "place": [1, 25, 46, 62, 77, 88, 90], "placement": 24, "plai": 74, "plan": [3, 5, 24, 60], "planner": 88, "platform": [27, 28, 38, 41, 42, 44, 45, 60, 66, 68, 88, 89], "pleas": [2, 5, 7, 10, 12, 19, 21, 22, 23, 24, 28, 32, 40, 60, 62, 68, 69, 71, 73, 77, 87, 88, 89, 94], "plu": 82, "plugin": [5, 6, 7, 11, 13, 59, 60, 65, 74, 77, 79, 83, 84, 85, 87, 88], "plugin_config": [65, 72, 75, 77, 79], "plugin_namespac": 7, "plugin_typ": 7, "plugin_v2": 7, "plugin_v2_gemm_0": 87, "pluginconfig": [65, 80], "pluginconfigmeta": 80, "pluginfield": 88, "pluginv2build": 87, "pm": [18, 24, 68], "pmi": 87, "pmi2_init": 87, "pmix": [14, 26, 50, 51, 52, 87], "png": [30, 35, 56], "po": 78, "point": [1, 5, 14, 16, 20, 23, 36, 40, 45, 61, 62, 64, 65, 69, 71, 76, 77, 83, 85, 87, 88], "pointer": [0, 1, 6, 15, 77, 82, 88], "pointerelementtyp": 1, "polar": 86, "polici": [0, 1, 2, 65, 68, 70, 84], "poll": [0, 26], "polyhedr": 14, "pong": 88, "pool": [0, 1, 5, 59, 77, 82, 93, 94], "pooled_project": [78, 79], "pooled_projection_dim": 78, "pooledpin": 0, "poor": 2, "popd": 87, "popfirstgentoken": 0, "popul": [1, 5, 14, 54, 77], "popular": [5, 13, 17, 23, 28, 64], "port": [0, 26, 28, 33], "portfolio": 21, "portion": [4, 71, 77, 84], "pos_emb_typ": 77, "pos_embd_param": 92, "pos_embed_max_s": 78, "pos_embed_typ": 78, "pose": 75, "posit": [0, 1, 10, 24, 65, 68, 77, 78, 82, 88, 92], "position_embed": [77, 78], "position_embedding_typ": [5, 13, 77, 78, 79], "position_encoding_2d": 79, "position_id": [79, 82, 87, 90, 92], "positionalembeddingparam": 92, "positionembeddingtyp": [5, 77, 78, 79], "positionid": [0, 1], "positionidsbas": 1, "positionidsdevic": 1, "positionidshost": 1, "positionidshostcopi": 1, "positionoffset": 1, "positionoffsetsdevic": 1, "positionoffsetshost": 1, "positionoffsetshostcopi": 1, "possibl": [2, 3, 5, 6, 8, 10, 14, 18, 25, 32, 60, 66, 67, 68, 69, 72, 74, 76, 77, 84, 87, 88, 91], "possibli": [1, 77], "post": [0, 13, 20, 23, 24, 54, 66, 67, 77, 83, 88], "post_act_fn": 78, "post_attention_layernorm": [15, 90], "post_input_id": 82, "post_layernorm": [12, 13, 15, 77, 87], "post_pad": 77, "post_prompt": 82, "post_strid": 77, "posterior_threshold": [39, 65], "posterioralpha": 1, "posterioralphahost": 1, "posteriorthreshold": [0, 1], "posteriorthresholdhost": 1, "postprocess": [26, 78], "postprocessor": [0, 65], "postprocparam": 65, "potenti": [0, 1, 10, 25, 67, 68, 72, 90], "pow": 77, "power": [8, 14, 21, 23, 24, 66, 74, 88], "pp": [0, 2, 6, 9, 19, 22, 26, 68, 70, 77, 88], "pp2": 68, "pp_communicate_final_output_id": 82, "pp_communicate_new_token": 82, "pp_reduce_scatt": [25, 75], "pp_size": [13, 14, 26, 33, 68, 69, 71, 81, 88], "ppreducescatt": 1, "pr": 24, "practic": [5, 14, 20, 21, 24, 83, 84, 88], "pre": [0, 1, 3, 5, 13, 16, 60, 62, 65, 66, 68, 77, 83, 84, 88, 92], "pre_input_id": 82, "pre_layernorm": 77, "pre_onli": 78, "pre_pad": 77, "pre_prompt": 82, "pre_quant_scal": [13, 65], "pre_strid": 77, "prebuilt": 60, "preced": [14, 77], "precis": [1, 6, 15, 19, 23, 25, 59, 68, 72, 75, 80, 83, 84, 86, 88], "precompute_relative_attention_bia": 79, "precomputed_relative_attent": 78, "predefin": [10, 90, 92], "predict": [1, 5, 10, 24, 88], "predicteddraftlogit": 1, "predictor": 10, "predictsdrafttoken": 1, "prefer": [23, 60], "prefer_managed_weight": 78, "prefer_plugin": 77, "prefetch": 24, "prefil": [0, 65, 73], "prefix": [3, 10, 13, 46, 64, 70, 77, 80, 87], "preliminari": [19, 21, 22], "preload": 15, "prepar": [0, 1, 2, 24, 46, 51, 59, 67, 74, 77, 79, 85, 88, 92], "prepare_dataset": [18, 51, 67, 68, 69, 70], "prepare_input": [79, 84], "prepare_position_ids_for_cogvlm": 82, "prepare_recurrent_input": 79, "prepare_resourc": [91, 93], "prepareforward": 1, "prepend": 87, "preprocess": [15, 82, 85], "preprocess_weights_hook": 79, "preprocessor": 68, "prequant_scaling_factor": 13, "prerequisit": [59, 61, 62], "presenc": [6, 14, 46], "presence_penalti": [65, 82, 88], "presencepenalti": [0, 1, 6], "present": [0, 65, 68, 74, 75, 85, 88], "preserv": 72, "presid": [36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 54, 61, 62, 70, 76, 83, 89], "pretrain": 16, "pretrained_config": 90, "pretrained_model_name_or_path": 79, "pretrainedconfig": [12, 17, 65, 79, 80, 90], "pretrainedmodel": [17, 79, 84], "pretrainedtokenizerbas": 65, "prevdrafttokenslen": 1, "prevent": [24, 59, 64], "preview": 88, "previou": [1, 3, 4, 10, 17, 18, 20, 68, 70, 71, 72, 74, 75, 76, 88], "previous": [1, 19, 72, 74, 76, 88], "prevscor": 1, "prewritten": 83, "price": 68, "primari": [0, 1, 23, 84, 94], "primarili": 92, "primit": [14, 66, 83], "print": [1, 5, 26, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 55, 56, 57, 61, 62, 68, 69, 70, 76, 83, 84, 87, 89], "print_iter_log": [18, 51], "prior": [3, 25, 60, 62], "priorit": [23, 74, 76], "prioriti": [0, 1, 8, 15, 65], "prioritytyp": 0, "priorityupd": 0, "privat": [0, 1, 6], "privileg": 7, "prm": 86, "pro": 24, "prob": 77, "probabilist": 78, "probabl": [0, 1, 6, 8, 10, 24, 65, 77, 82, 88], "probil": 1, "problem": [5, 18, 87], "proc": 15, "proccessed_weight": 15, "proccessed_zero": 15, "procedur": 18, "proceed": 14, "process": [0, 1, 2, 3, 5, 6, 10, 13, 14, 17, 18, 24, 25, 36, 40, 45, 47, 50, 51, 52, 61, 62, 64, 65, 66, 67, 68, 69, 70, 71, 74, 75, 76, 77, 82, 83, 87, 88, 90, 91, 92, 94], "process_input": 82, "process_logits_including_draft": 82, "processor": [0, 5, 36, 37, 48, 65, 79, 82, 88], "processorbatch": 0, "processormap": 0, "prod": 77, "produc": [0, 1, 3, 7, 14, 32, 68, 70, 72, 74, 75, 77, 88], "product": [4, 5, 10, 14, 21, 66, 74, 75, 76, 77, 83, 92], "profil": [2, 25, 26, 34, 35, 59, 72, 74, 77, 82, 84, 87, 88], "profiling_verbos": [25, 65], "profit": [10, 68], "program": [2, 17, 36, 38, 41, 42, 44, 45, 47, 61, 62, 64, 76, 83, 87], "progress": [1, 24, 65, 68, 77], "proj": [13, 15, 87], "project": [5, 9, 54, 60, 77, 78, 90, 93], "projector_hidden_act": 79, "prologu": [50, 51, 52], "promin": 10, "promis": [10, 17], "prompt": [0, 3, 6, 8, 12, 18, 25, 26, 31, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 53, 54, 57, 59, 61, 62, 65, 68, 70, 74, 75, 76, 78, 82, 83, 88, 89, 92], "prompt_adapter_request": [65, 88], "prompt_embedding_t": [78, 79, 82], "prompt_embedding_table_s": 79, "prompt_id": 47, "prompt_len": 92, "prompt_logprob": 65, "prompt_lookup": [10, 88], "prompt_lookup_num_token": 6, "prompt_tabl": 82, "prompt_task": [79, 82], "prompt_token": 83, "prompt_token_id": [32, 48, 65], "prompt_vocab_s": [79, 82], "promptadapterrequest": 65, "promptinput": [65, 88], "promptlen": 0, "prompttableoffload": 0, "prompttuningconfig": 0, "prompttuningembed": 78, "prompttuningen": 1, "pronounc": 10, "proof": 93, "propag": [8, 88], "proper": [2, 68], "properli": [15, 74, 76], "properti": [3, 40, 65, 77, 79, 80, 82], "proport": 5, "propos": 24, "protect": [1, 36, 45, 61, 62, 64, 83], "protocol": [0, 26, 40], "proud": 24, "prove": 10, "provid": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 16, 17, 18, 19, 20, 23, 24, 25, 26, 27, 32, 40, 49, 54, 60, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 77, 82, 84, 86, 87, 88, 90, 91, 92], "proxy_dispatch_result_thread": 68, "prune": [7, 10, 77], "pseudo": [5, 77, 85], "pth": [15, 88], "ptq": [23, 72, 88], "ptr": 1, "ptr_idx": 15, "ptrdiff_t": 1, "ptuning_setup": 82, "ptuning_setup_fuyu": 82, "ptuning_setup_llava_next": 82, "ptuning_setup_phi3": 82, "ptuning_setup_pixtr": 82, "ptuningconfig": 0, "public": [0, 1, 23, 28, 49, 54], "publish": [18, 19, 22, 68, 69, 88], "pull": [16, 18, 60, 83, 88], "puneeshkhanna": 88, "purchas": 68, "pure": 82, "purpos": [5, 60, 70, 72, 74, 75], "pursu": [38, 41, 42, 44, 45, 47], "push": [27, 48], "pushd": 87, "put": [1, 13, 24, 50, 51, 52, 64, 66, 74], "pwd": [18, 60], "py": [3, 4, 5, 7, 9, 10, 11, 12, 13, 14, 15, 17, 18, 24, 47, 50, 51, 60, 62, 64, 67, 68, 69, 70, 71, 72, 77, 80, 82, 83, 87, 88, 90, 91, 93, 94], "py3": 88, "py_executor_cr": 94, "pybind": 88, "pybind11_object": 65, "pybindmirror": 65, "pydant": [65, 88], "pydantic_cor": 65, "pyexecutor": [46, 88, 93, 94], "pynvml": 88, "pypi": [60, 88], "python": [1, 5, 6, 7, 9, 10, 12, 14, 16, 17, 18, 26, 32, 42, 43, 59, 61, 62, 64, 67, 68, 69, 70, 71, 83, 85, 88, 90, 91, 93, 94], "python3": [9, 11, 13, 18, 50, 51, 60, 62, 67, 68, 83, 87], "python_bind": 18, "python_e2": 82, "python_plugin": 88, "pythonpath": [18, 51, 52], "pytorch": [7, 10, 13, 16, 18, 26, 33, 46, 50, 51, 52, 59, 60, 61, 62, 65, 69, 77, 88, 91, 92, 93, 94], "pytorch_backend_config": [18, 26, 46, 51, 68, 69, 92], "pytorch_config": [46, 92], "pytorch_eagle_weights_path": 65, "pytorch_extra_arg": 51, "pytorch_model": 87, "pytorch_model_engin": 91, "pytorch_model_registri": 93, "pytorchconfig": [46, 92], "pytorchmodelengin": [91, 93], "pzzzzz5142": 88, "q": [2, 5, 6, 9, 19, 24, 59, 68, 77, 87, 90, 92], "q_b_proj": 77, "q_dim": 77, "q_lora_rank": [77, 78], "q_proj": [15, 90], "q_scale": [5, 77, 78, 79], "qa": 10, "qformat": [68, 81], "qgmma": 88, "qingquansong": 88, "qk_layernorm": [78, 79], "qk_nope_head_dim": [77, 78], "qk_norm": 78, "qk_rope_head_dim": [77, 78], "qkv": [7, 9, 13, 15, 59, 77, 87, 88, 92], "qkv_bia": [77, 88], "qkv_dim": 77, "qkv_proj": 90, "qo_indptr": 92, "qserv": 88, "quadrat": [5, 84], "qualiti": [72, 75], "qualnam": [65, 77, 79, 81], "quant": [17, 65, 68, 77, 88, 89], "quant_algo": [13, 15, 17, 32, 54, 65, 68, 72, 79], "quant_and_calib_config": 54, "quant_config": [17, 32, 54, 65, 72, 79, 92], "quant_medusa_head": 81, "quant_mod": [17, 65, 78, 79, 82], "quantalgo": [32, 54, 65, 72, 79, 81], "quantconfig": [17, 32, 54, 65, 72, 79, 88, 92], "quanticonfig": 17, "quantiz": [5, 6, 14, 15, 18, 19, 20, 24, 25, 36, 37, 41, 49, 59, 62, 63, 64, 65, 66, 69, 70, 73, 77, 78, 79, 82, 83, 86, 88, 90, 92], "quantizaton": 68, "quantize_and_export": 81, "quantize_kwarg": 79, "quantize_lm_head": [81, 88], "quantized_valu": 5, "quantizedkernel": 14, "quantizetensorplugin": 14, "quantmod": [1, 5, 6, 59, 65, 77, 78, 79, 81, 82], "quantmodewrapp": [65, 77], "queri": [3, 6, 10, 14, 19, 26, 59, 68, 77, 84, 92, 93], "query_dim": 78, "query_key_valu": 15, "query_length": 78, "query_pre_attn_scalar": 79, "question": [53, 68, 84, 87], "queu": [0, 69, 74], "queue": [0, 65, 66, 91], "quick": [5, 59, 66, 68, 70, 92], "quick_gelu": 77, "quicker": 71, "quickli": [17, 83], "quickstart": [64, 70], "quickstart_advanc": 50, "quit": [7, 64], "qweight": 15, "qwen": [15, 26, 35, 64, 68, 77, 85, 86, 88], "qwen1": [86, 88], "qwen2": [9, 26, 30, 35, 56, 68, 86, 88], "qwen2_5_vlforconditionalgener": 86, "qwen2audio": 88, "qwen2forcausallm": 86, "qwen2forprocessrewardmodel": 86, "qwen2forrewardmodel": 86, "qwen2forsequenceclassif": 88, "qwen2vl": 88, "qwen2vlforconditionalgener": 86, "qwenforcausallm": 15, "qwenforcausallmgenerationsess": 82, "qwenvl": 88, "qwq": 86, "qychen": 9, "qzero": 15, "r": [1, 9, 26, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 61, 62, 70, 76, 77, 83, 87, 88, 89], "r1": [26, 58, 69, 88], "r1_in_tensorrt": [24, 88], "race": 88, "radix": 93, "rai": 1, "rais": [17, 65, 70, 87, 88], "rand": 77, "rand_data": 77, "rand_data_sampl": 79, "rand_data_valid": 79, "random": [0, 6, 26, 34, 35, 65, 69, 77, 88], "random_se": [65, 79, 82], "randomdatasampl": 1, "randomdatavalid": 1, "randomli": 69, "randomse": [1, 6, 88], "randomseedtyp": 0, "rang": [0, 6, 8, 10, 67, 75, 77, 79, 84, 85, 86, 87, 90], "rank": [0, 1, 2, 3, 4, 6, 9, 17, 18, 25, 64, 68, 77, 79, 82, 84, 87, 88], "rank0": 13, "rank1": 13, "rapid": [10, 69, 83], "rate": [0, 18, 24, 26, 34, 35, 68, 69, 70, 88], "rather": [5, 7, 10, 62, 66], "raw": 26, "raw_audio": 82, "raw_imag": 82, "rdma": 2, "re": [18, 23, 65, 66, 88, 92], "reach": [0, 5, 13, 64, 68, 72, 76], "read": [2, 3, 5, 10, 12, 14, 15, 18, 24, 25, 53, 68, 88], "read_config_from_the_custom_training_checkpoint": 17, "readabl": 68, "reader": 77, "readi": [0, 83], "readm": [2, 10, 26, 64, 70, 88], "real": [7, 18, 24, 60, 70, 72, 74, 75, 77, 87], "realiti": 74, "realiz": [8, 10], "rearrang": 77, "reason": [0, 5, 6, 14, 17, 24, 26, 58, 65, 68, 71, 74, 75, 77, 87], "reasoning_pars": [26, 33], "rebuild": [75, 77, 87], "receiv": [0, 1, 2, 3, 4, 10, 72, 77, 88], "recent": [1, 4, 5, 20, 24], "recip": [24, 26, 65, 85], "reclaim": 0, "recogn": [10, 24, 68, 90], "recommend": [2, 5, 6, 10, 12, 15, 16, 18, 20, 23, 26, 47, 60, 65, 68, 73, 74, 76, 87, 88, 90, 92], "recompute_scale_factor": 77, "reconfigur": [3, 62], "reconstruct": [5, 77], "record": [1, 7, 18, 24, 65], "recored": 0, "recreat": 16, "recurr": 10, "recurrentgemma": [85, 86, 88], "recurrentgemmaforcausallm": 79, "recurs": [18, 60, 64], "recv": [0, 14, 77], "recvconnect": 0, "recvpollperiodm": 0, "recycl": [5, 93], "redesign": 88, "redirect": [7, 65], "redraft": [59, 77, 82, 88], "redrafter_draft_len_per_beam": 82, "redrafter_inverted_temperatur": 79, "redrafter_num_beam": 82, "redrafterforcausallm": 79, "reduc": [2, 3, 4, 5, 8, 10, 14, 18, 19, 22, 24, 25, 60, 64, 66, 67, 68, 69, 70, 71, 74, 76, 77, 84, 87, 88, 92], "reduce_fus": [25, 68, 72, 75], "reduce_scatt": 77, "reduceoper": 77, "reducescatt": [25, 75, 88], "reduct": [10, 24, 76, 77], "redund": [10, 24], "refactor": [17, 88], "refer": [0, 1, 2, 3, 5, 6, 7, 9, 10, 14, 16, 17, 18, 26, 28, 29, 30, 31, 32, 33, 34, 35, 36, 47, 55, 56, 57, 60, 64, 66, 68, 69, 70, 71, 72, 73, 75, 77, 83, 86, 88, 90, 92], "referenc": 72, "reference_wrapp": [0, 3], "refin": 88, "refit": [14, 25, 88], "refit_engin": 14, "reflect": 74, "refresh": 68, "regard": 77, "regardless": 87, "regex": [3, 65], "region": 67, "regist": [27, 59, 87, 88, 90], "register_auto_model": 90, "register_network_output": 87, "regress": [5, 6, 14], "regular": [0, 3, 5, 24, 65, 77], "reinforc": 73, "reject": 0, "rel": [8, 19, 74, 76, 77, 88], "rel_attn_t": 78, "relat": [2, 4, 15, 59, 66, 67, 77, 80, 84, 87, 88, 89, 90, 93], "relationship": 84, "relative_attent": [77, 78], "relative_attention_bia": 77, "relax": 5, "relaxed_delta": [24, 65], "relaxed_topk": [24, 65], "releas": [1, 5, 6, 17, 19, 22, 23, 59, 66, 77, 79, 84, 85, 86], "release_build": 60, "release_run": [60, 83], "releasepag": 1, "releasest": 0, "relev": [6, 60, 93], "reli": [2, 5, 7, 17, 64, 67, 85], "reload": 3, "relu": [13, 14, 77, 87], "remain": [0, 7, 8, 10, 11, 24, 60, 69, 70, 72, 74, 75, 77, 84, 88], "remaind": 72, "remark": 24, "remind": [5, 92], "remot": 65, "remov": [0, 1, 5, 6, 7, 14, 15, 18, 25, 26, 49, 60, 65, 66, 72, 77, 84, 88, 90], "remove_const_t": 1, "remove_cv_t": 0, "remove_duplicated_kv_head": 79, "remove_input_pad": [5, 9, 25, 77, 78, 82], "remove_pointer_t": 1, "remove_reference_t": 1, "remove_sequ": 93, "renam": 88, "reorder": [77, 78], "reorder_kv_cache_for_beam_search": 82, "rep": 67, "repeat": [0, 5, 65, 77], "repeat_interleav": 77, "repeatedli": 10, "repetit": [0, 6, 65, 77], "repetition_penalti": [6, 65, 82, 88], "repetitionpenalti": [0, 1, 6], "replac": [1, 4, 7, 14, 15, 17, 18, 68, 70, 72, 76, 77, 84, 90], "replace_add_with_sub": 7, "replace_all_uses_with": [7, 77], "replace_input_with": 7, "replace_output_uses_with": 7, "replace_outputs_uses_with": 7, "replic": [0, 3, 24, 77], "replit": [85, 86, 88], "repo": [17, 64, 66, 70, 87], "repo_id": 53, "report": [67, 68, 69, 84, 88], "reportpluginerror": 87, "repositori": [10, 16, 18, 27, 64, 83], "repres": [0, 1, 2, 10, 18, 19, 23, 24, 40, 53, 65, 68, 74, 77, 82, 94], "represent": [7, 14], "reproduc": [59, 68, 88], "req": [18, 68, 69, 70, 72, 74, 75], "req_id": 47, "req_logit": 47, "req_stat": 94, "req_token_id": 47, "reqbeamwidth": 1, "reqid": 0, "reqpromptlength": 1, "request": [0, 2, 5, 6, 8, 9, 14, 18, 20, 22, 25, 26, 34, 35, 47, 51, 65, 66, 67, 68, 69, 70, 72, 74, 75, 76, 77, 83, 84, 88, 91, 92, 93, 94], "request_id": [32, 48, 65, 92], "request_stats_max_iter": 65, "request_timeout": 26, "request_typ": 65, "request_type_context_and_gener": [0, 2], "request_type_context_onli": [0, 2], "request_type_generation_onli": [0, 2], "requesterror": 65, "requestid": [0, 2, 3], "requestidtyp": 0, "requestlist": 94, "requestoutput": [32, 48, 65, 88], "requestperfmetr": 0, "requestschedul": 94, "requeststag": 0, "requeststat": 0, "requeststatsmaxiter": 0, "requeststatsperit": 0, "requeststatsperiter": 0, "requeststatsvec": 0, "requesttoken": 3, "requesttyp": [0, 1, 2, 65], "requesttypesdevic": 1, "requestvector": 1, "requir": [0, 2, 5, 6, 8, 9, 10, 14, 15, 17, 18, 19, 23, 24, 25, 26, 40, 53, 60, 61, 62, 65, 68, 69, 70, 71, 72, 75, 77, 78, 83, 84, 86, 87, 88, 93], "require_ln_f": 79, "requiresattentionmask": 1, "rerun": 75, "rescale_output_factor": 78, "research": [5, 28, 38, 41, 42, 44, 45, 85], "resembl": 46, "reserv": [0, 1, 26, 65, 76, 82, 84, 94], "reserved_block": 94, "reset": [0, 1, 6, 65, 68, 82], "resetspeculativedecodingmodul": 1, "reshap": [1, 77], "resid": [9, 54], "residu": [77, 87], "residual_connect": 78, "residual_mlp": 79, "residual_multipli": 79, "residual_rms_norm": 77, "residual_rms_norm_out_quant_fp8": 77, "residual_rms_norm_out_quant_nvfp4": 77, "residual_rms_norm_quant_fp8": 77, "residual_rms_norm_quant_nvfp4": 77, "residual_rms_prepost_norm": 77, "residualadd": [25, 75, 88], "resiz": 1, "resolv": [30, 56, 87], "resourc": [0, 2, 5, 17, 24, 91, 93, 94], "respect": [4, 32, 76, 77, 82, 84, 85, 90, 94], "respons": [0, 2, 26, 32, 55, 56, 57, 65, 68, 77, 91], "responsewithid": 0, "rest": [1, 5, 72], "restart": 0, "restrict": [0, 2, 3, 6, 60, 65, 77], "result": [0, 1, 4, 5, 10, 14, 19, 20, 21, 23, 25, 32, 59, 60, 65, 68, 71, 72, 73, 74, 75, 77, 78, 88, 90, 92, 94], "retail": 68, "retain": [19, 21], "retent": [0, 65], "retentionprior": 0, "retentionpriorityanddur": 0, "rethink": 10, "retriev": [1, 15, 65, 69, 77], "return": [0, 1, 3, 7, 9, 10, 12, 14, 15, 17, 32, 65, 68, 74, 77, 78, 79, 82, 84, 87, 88, 93, 94], "return_all_generated_token": 82, "return_context_logit": 65, "return_dict": 82, "return_encoder_output": [65, 82], "return_generation_logit": 65, "return_perf_metr": 65, "returnallgeneratedtoken": [0, 3], "returncontextlogit": 0, "returnencoderoutput": 0, "returngenerationlogit": 0, "returnlogprob": 0, "returnperfmetr": 0, "reus": [0, 2, 3, 25, 59, 63, 65, 77, 82, 84, 88, 90, 93], "reusabl": 8, "reusedblock": 0, "reusedblocksperrequest": 0, "reveal": 24, "revers": 77, "revert": 77, "review": 68, "revis": 65, "revolution": 66, "rewind": 88, "rewrit": [59, 77, 88, 90], "rewritepatternmanag": 7, "rewrt": 87, "rf": 87, "rg_lru": 77, "rgc": 68, "rh": [0, 1], "rich": 13, "right": [66, 72, 77, 87], "rigor": [46, 68], "risk": [2, 14, 72, 76], "rm": [60, 77, 86, 87, 90], "rms_norm": [24, 77, 90], "rmsnorm": [9, 24, 77, 78, 79, 88, 90], "rnn": [25, 88], "rnn_conv_dim_s": 82, "rnn_head_siz": 82, "rnn_hidden_s": 82, "rnn_state": 79, "rnnconfig": 1, "rnnconvdims": 1, "rnnheadsiz": 1, "rnnhiddens": 1, "ro": 18, "roberta": [86, 88], "robertaforquestionansw": 79, "robertaforsequenceclassif": 79, "robertamodel": 79, "robin": 2, "robust": [24, 88], "rock": 77, "role": [14, 26, 29, 30, 40, 55, 56, 74, 83], "roll": 59, "root": [13, 18, 27, 60, 62, 64, 65, 70, 77, 83], "root_lay": 7, "rope": [24, 77, 82, 88, 92], "rope_gpt_neox": [5, 77, 79], "rope_gptj": [5, 77], "rope_local_base_freq": 79, "rope_scaling_config": 77, "rope_scaling_long_factor": 78, "rope_scaling_long_mscal": 78, "rope_scaling_short_factor": 78, "rope_scaling_short_mscal": 78, "ropeembeddingutil": 77, "rotari": [0, 24, 77, 82, 90, 92], "rotary_bas": 79, "rotary_cos_sin": 77, "rotary_dim": 79, "rotary_embed": 90, "rotary_embedding_bas": [77, 78], "rotary_embedding_base_loc": 78, "rotary_embedding_beta_fast": 78, "rotary_embedding_beta_slow": 78, "rotary_embedding_dim": [5, 77, 79], "rotary_embedding_long_m_scal": 77, "rotary_embedding_max_posit": 77, "rotary_embedding_mscal": 78, "rotary_embedding_mscale_all_dim": 78, "rotary_embedding_origin_max_posit": 78, "rotary_embedding_original_max_posit": 77, "rotary_embedding_percentag": 78, "rotary_embedding_sc": 78, "rotary_embedding_scal": 77, "rotary_embedding_scale_typ": 77, "rotary_embedding_short_m_scal": 77, "rotary_inv_freq": [77, 78], "rotary_inv_freq_loc": 78, "rotary_pct": 79, "rotary_sc": [78, 79], "rotaryembed": 90, "rotaryembeddingdim": [0, 1], "rotaryscalingtyp": 77, "rotate_every_two": 77, "rotate_half": 77, "round": [2, 77], "rout": 2, "router": [4, 9, 88], "router_gemm": 24, "routin": 7, "routingkernel": 24, "row": [9, 74, 77, 85, 88], "rowlinear": [9, 78], "rowwis": 65, "rr": 88, "rslora": 88, "rst": 3, "rtx": 88, "rubric": 77, "rule": [5, 71, 87], "run": [0, 1, 2, 3, 5, 6, 8, 10, 12, 13, 14, 19, 23, 24, 25, 26, 27, 28, 42, 43, 47, 50, 51, 52, 59, 60, 61, 62, 64, 65, 66, 71, 72, 74, 75, 76, 77, 79, 82, 84, 85, 87, 88, 90, 91, 92, 93], "run_dtm_pld": 10, "run_medusa_decod": 49, "runner": [0, 13, 82], "runningleon": 88, "runpod": 27, "runtim": [0, 3, 5, 10, 11, 16, 24, 25, 26, 44, 47, 53, 59, 65, 66, 67, 68, 70, 73, 74, 77, 78, 79, 83, 87, 88, 90, 92, 94], "runtime_config": [32, 44], "runtime_default": 79, "runtime_error": 1, "runtime_rank": 82, "runtimedefault": [0, 79], "runtimedefaultsin": 79, "runtimeerror": [64, 65, 87], "runtimetensor": 82, "s0": 5, "s1": 5, "s2": 5, "sacrif": 24, "sad": 82, "saeyoonoh": 88, "safe": [1, 7, 75], "safer": 77, "safetensor": [13, 15, 87, 88], "sage_attn": 77, "sage_attn_k_block_s": 77, "sage_attn_k_quant_s": 77, "sage_attn_q_block_s": 77, "sage_attn_q_quant_s": 77, "sage_attn_v_block_s": 77, "sage_attn_v_quant_s": 77, "sageattent": 77, "sai": [67, 70, 74], "said": 72, "sake": 74, "sale": 68, "same": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 14, 17, 20, 25, 47, 50, 51, 52, 60, 64, 68, 69, 72, 75, 76, 77, 78, 80, 82, 84, 88], "sampl": [0, 1, 3, 5, 14, 16, 18, 24, 39, 41, 42, 43, 44, 45, 46, 47, 49, 53, 54, 59, 63, 65, 67, 68, 69, 77, 78, 82, 88], "sample_proj_bia": 78, "sample_weight_strip": 88, "samplemod": 77, "sampling_config": 82, "sampling_param": [32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 54, 61, 62, 65, 70, 76, 83, 88, 89], "samplingconfig": [0, 3, 6, 32, 82, 88], "samplingparam": [32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 54, 61, 62, 65, 70, 76, 83, 88, 89], "saniti": [61, 62, 71, 72, 75], "santacod": [64, 85, 86], "satfinit": 85, "satisfi": [6, 15, 88], "save": [5, 8, 10, 17, 18, 25, 27, 41, 44, 64, 65, 67, 68, 72, 75, 76, 84, 88], "save_checkpoint": [17, 79], "save_config": [17, 79], "saw": [72, 83], "sbatch": [14, 50, 51, 52], "sbsa": [88, 89], "scaffold": [88, 90], "scalar": [6, 77], "scalartyp": 88, "scale": [0, 6, 9, 15, 25, 65, 72, 77, 78, 85, 88], "scale_d0": 77, "scale_d1": 77, "scale_factor": 77, "scale_output": 77, "scale_qk": 78, "scale_typ": 77, "scalia": [38, 41, 42, 44, 45], "scaling_factor": 77, "scaling_long_factor": 77, "scaling_short_factor": 77, "scalingvecpoint": 1, "scanreducetempstorag": 1, "scanreducetempstoragebyt": 1, "scantempstorag": 1, "scantempstoragebyt": 1, "scatter": [7, 77], "scatter_nd": 77, "scenario": [2, 5, 10, 13, 18, 21, 23, 24, 25, 28, 68, 69, 70, 72, 74, 75, 88], "scfg": 82, "schedul": [0, 2, 3, 8, 9, 18, 25, 26, 46, 65, 68, 70, 75, 84, 88, 89], "schedule_request": 94, "scheduled_request": 94, "scheduler_config": [65, 76], "schedulerconfig": [0, 65, 76, 88], "schedulerpolici": 88, "schema": [0, 3, 40, 65, 68], "scheme": 0, "scicod": 24, "scienc": [38, 41, 42, 44, 45, 47], "scope": [16, 88], "score": 6, "scout": 86, "scratch": [68, 70, 71, 75], "script": [9, 12, 14, 17, 18, 27, 50, 51, 52, 60, 64, 67, 68, 69, 70, 80, 85, 87, 88, 89, 90], "sd3": 78, "sd35adalayernormzerox": 78, "sd3patchemb": 78, "sd3transformer2dmodel": 79, "sd3transformer2dmodelconfig": 79, "sdxl": 88, "seamless": 88, "search": [0, 1, 3, 6, 10, 16, 22, 25, 26, 32, 44, 59, 65, 72, 74, 77, 88, 91], "seashor": [30, 56], "seat": [38, 41, 42, 44, 45], "sec": [18, 20, 68, 69, 70, 72, 74, 75], "second": [1, 3, 6, 8, 9, 10, 18, 19, 21, 22, 24, 65, 74, 77], "secondari": [0, 65, 84], "secondary_offload_min_prior": 65, "secondaryoffloadminprior": 0, "secondli": 74, "section": [3, 6, 14, 15, 17, 18, 26, 60, 64, 66, 68, 70, 72, 73, 74, 75, 77, 83, 86, 88, 92], "section_s": 77, "secur": [40, 88], "securityprotocol": 40, "see": [0, 1, 5, 6, 10, 14, 15, 18, 19, 21, 22, 23, 26, 27, 28, 30, 36, 56, 62, 68, 69, 70, 72, 74, 75, 76, 77, 78, 79, 84, 85, 87, 88], "seed": [0, 6, 26, 34, 35, 65, 81, 88], "seem": [8, 46, 53, 68, 71], "seen": [10, 18, 68], "segment": 88, "select": [0, 4, 6, 16, 23, 24, 25, 68, 75, 77, 82, 84, 91, 94], "selectcontextid": 0, "selectgenidx": 0, "selective_scan": 77, "self": [0, 5, 7, 12, 14, 15, 47, 65, 68, 77, 79, 82, 87, 90, 93, 94], "self_attent": 15, "self_attention_mask": 78, "self_attention_packed_mask": 78, "self_attn": [15, 90], "selfidx": 0, "sell": 68, "semicolon": 60, "senat": [38, 41, 42, 44, 45], "send": [0, 2, 14, 24, 26, 70, 71, 77, 83, 88], "sens": 72, "sensit": [24, 72], "sent": [0, 10, 26], "sentenc": [0, 6, 65, 83], "separ": [10, 25, 49, 60, 68, 77, 82, 92], "separate_match_rewrit": 7, "seq": [1, 5, 68, 77], "seq_idx": 82, "seq_len": [69, 77, 78, 92], "seq_length": 77, "seq_lens_cuda": 92, "seqlen": [0, 77], "seqslot": 1, "sequenc": [0, 1, 3, 5, 6, 7, 8, 10, 14, 18, 19, 20, 21, 22, 24, 65, 66, 68, 69, 70, 73, 76, 77, 78, 82, 84, 88, 92, 93], "sequence_length": [77, 78, 82, 87], "sequence_length_buff": 82, "sequence_limit_length": 82, "sequenceindex": [0, 3], "sequencelengthscba": 1, "sequencelimitlength": 1, "sequenti": [0, 2, 10, 84], "seri": 88, "serial": [25, 77, 79, 82], "serializ": 65, "serialize_engin": 82, "serializeds": 0, "serializedst": 0, "serv": [0, 2, 3, 5, 10, 14, 16, 22, 23, 29, 30, 31, 33, 34, 35, 36, 37, 55, 56, 57, 59, 65, 75, 88, 91, 92], "server": [0, 8, 10, 14, 16, 20, 27, 29, 30, 31, 33, 34, 35, 55, 56, 57, 59, 88], "server_start_timeout": 26, "servic": [16, 54, 59], "session": [5, 64, 68, 82], "set": [0, 1, 2, 3, 4, 5, 6, 7, 10, 11, 13, 15, 16, 17, 18, 24, 25, 26, 32, 40, 50, 51, 52, 60, 62, 65, 66, 67, 69, 70, 72, 74, 75, 76, 77, 78, 79, 80, 82, 83, 84, 87, 88, 94], "set_attn_processor": 79, "set_from_opt": 1, "set_if_not_exist": 79, "set_input_shap": 82, "set_rank": 79, "set_rel_attn_t": 78, "set_shap": 82, "setadditionalmodeloutput": [0, 3], "setallottedtimem": 0, "setbackend": 0, "setbadword": 0, "setbatchingtyp": 0, "setbeamsearchdiversityr": 0, "setbeamwidth": 0, "setbeamwidtharrai": 0, "setbitto": 0, "setcachest": 0, "setcachetransceiverconfig": 0, "setclientid": 0, "setcommst": 0, "setcommunicationmod": 0, "setcommunicationtyp": 0, "setcontextfmha": 1, "setcontextphaseparam": [0, 2], "setcopyonpartialreus": 0, "setcrossattentionmask": 0, "setcrosskvcachefract": 0, "setcudagraphcaches": 0, "setcudagraphmod": 0, "setdatatyp": 1, "setdebugconfig": 0, "setdebuginputtensor": 0, "setdebugoutputtensor": 0, "setdebugtensornam": 0, "setdebugtensorsmaxiter": 0, "setdecodingconfig": 0, "setdecodingmod": 0, "setdeviceid": 0, "seteagleconfig": 0, "seteagleinput": 1, "setearlystop": 0, "setembeddingbia": 0, "setenableblockreus": 0, "setenablechunkedcontext": 0, "setenablecontextfmhafp32acc": 0, "setenablepartialreus": 0, "setenabletrtoverlap": 0, "setencodedvocab": 0, "setencoderhiddens": 1, "setencoderinputfeatur": 0, "setencoderinputtokenid": 0, "setencoderoutputlength": 0, "setendid": 0, "seteventbuffermaxs": 0, "setexecutionconfig": 1, "setexplicitdrafttokensinput": 1, "setextendedruntimeperfknobconfig": 0, "setexternaldrafttokensconfig": 0, "setfreegpumemoryfract": 0, "setfrequencypenalti": 0, "setfrom": 0, "setfrominput": 1, "setgathergenerationlogit": 0, "setgemmallreducedtyp": 1, "setgpuweightsperc": [0, 11], "setguideddecodingconfig": 0, "setguideddecodingparam": 0, "sethostcaches": 0, "setinittozero": 1, "setisorchestr": 0, "setiterstatsmaxiter": 0, "setkvcacheconfig": 0, "setkvcacheretentionconfig": 0, "setkvcachetyp": 1, "setlanguageadapteruid": 0, "setlayertyp": 1, "setlengthpenalti": 0, "setlevel": 1, "setlogitsdtyp": 1, "setlogitspostprocessor": 0, "setlogitspostprocessorconfig": 0, "setlogitspostprocessornam": 0, "setlookaheadconfig": 0, "setlookaheaddecodingconfig": 0, "setloraconfig": 0, "setloramodul": 1, "setmanagedweightsmap": 1, "setmanageweightstyp": 1, "setmaxattentionwindowvec": 0, "setmaxbatchs": [0, 1], "setmaxbeamwidth": [0, 1], "setmaxdraftpathlen": 1, "setmaxdrafttoken": 1, "setmaxencoderlen": 1, "setmaxinputlen": 1, "setmaxlorarank": 1, "setmaxnumpath": 1, "setmaxnumtoken": [0, 1], "setmaxpagesperblock": 1, "setmaxpositionembed": 1, "setmaxpromptembeddingtables": 1, "setmaxqueues": 0, "setmaxseqidlemicrosecond": 0, "setmaxsequencelen": 1, "setmaxtoken": 0, "setmedusachoic": 0, "setmem": 1, "setmemorytyp": 1, "setminp": 0, "setmintoken": 0, "setmlphiddens": 1, "setmodelnam": 1, "setmodelvari": 1, "setmropeconfig": 0, "setmultiblockmod": 0, "setmultimodalembed": 0, "setnbcrosskvhead": 1, "setnbkvhead": 1, "setnorepeatngrams": 0, "setnormalizelogprob": 0, "setnumcopystream": 1, "setnumdecodingenginetoken": 1, "setnumkvheadspercrosslay": 1, "setnumkvheadsperlay": 1, "setnumlanguag": 1, "setnumnod": 0, "setnumreturnsequ": 0, "setonboardblock": 0, "setorchestratorconfig": 0, "setorchleadercomm": 0, "setoutputconfig": 0, "setpadid": 0, "setpagedcontextfmha": 1, "setpagewidth": 1, "setparallelconfig": 0, "setparticipantid": 0, "setpath": 1, "setpeftcacheconfig": 0, "setpositionid": 0, "setppreducescatt": 1, "setpresencepenalti": 0, "setprior": 0, "setprocessorbatch": 0, "setprocessormap": 0, "setprompttableoffload": 0, "setprompttuningconfig": 0, "setquantmod": 1, "setrecvpollperiodm": 0, "setrepetitionpenalti": 0, "setrepl": [0, 3], "setrequeststatsmaxiter": 0, "setrequesttyp": [0, 2], "setreturnallgeneratedtoken": 0, "setrnnconfig": 1, "setrotaryembeddingdim": 1, "setsamplingconfig": 0, "setschedulerconfig": 0, "setse": 0, "setsecondaryoffloadminprior": 0, "setsinktokenlength": 0, "setsizeperhead": 1, "setskipcrossattnblock": [0, 1], "setslotsperpag": 1, "setspawnprocess": 0, "setspecdecconfig": 0, "setspeculativedecodingmod": 1, "setspeculativedecodingmodul": 1, "setstoptokenid": 0, "setstopword": 0, "setstream": 0, "settemperatur": 0, "setter": [0, 6], "settokenizerstr": 0, "settokensperblock": 1, "settopk": 0, "settopp": 0, "settoppdecai": 0, "settoppmin": 0, "settoppresetid": 0, "settotalnumpag": 1, "setup": [1, 5, 25, 40, 50, 51, 52, 62, 71, 72, 82, 83, 84, 88], "setup_fake_prompt": 82, "setup_fake_prompts_qwen2vl": 82, "setup_fake_prompts_vila": 82, "setup_input": 82, "setupeagl": 1, "setupexplicitdrafttoken": 1, "setuplookahead": 1, "setupspeculativedecod": 1, "setuptool": [61, 62], "setusecrossattent": 1, "setusegpudirectstorag": 0, "setusemrop": 1, "setusepositionembed": 1, "setuseshapeinfer": 1, "setusetokentypeembed": 1, "setworkerexecutablepath": 0, "setzero": [0, 1], "seve": 65, "sever": [0, 1, 2, 5, 7, 10, 13, 32, 72, 73, 74, 75, 77, 84, 87, 92], "sft": 53, "sh": [14, 27, 88, 89], "shah": 88, "shaken": 46, "shall": [17, 84], "shape": [0, 1, 5, 7, 9, 13, 14, 24, 65, 75, 77, 79, 82, 84, 85, 87, 88, 92, 93], "shape_cast_dtyp": 77, "shapeequ": 1, "shard": [15, 24, 59, 68, 73, 77, 78], "shard_map": 15, "sharding_along_vocab": 65, "sharding_dim": [77, 78], "share": [1, 2, 3, 5, 7, 8, 9, 10, 17, 18, 23, 24, 25, 60, 71, 72, 77, 78, 88], "share_embed": 88, "share_weight": 78, "shared_embedding_t": 88, "shared_ptr": [0, 1], "sharedconstptr": 1, "sharedptr": 1, "shelf": 88, "sherlock113": 88, "ship": [17, 46], "shm": 87, "short": [5, 68, 72, 74], "short_mscal": [77, 78], "shorter": [5, 69], "shot": 88, "should": [0, 1, 2, 3, 7, 8, 9, 17, 18, 32, 38, 40, 41, 42, 44, 45, 47, 48, 50, 51, 52, 53, 60, 65, 68, 69, 70, 71, 75, 76, 77, 78, 80, 82, 84, 88, 90, 92, 93, 94], "should_stop": 82, "shouldus": 5, "show": [2, 3, 14, 20, 24, 26, 36, 69, 70, 74, 75, 83, 84, 86, 89], "showcas": [72, 75, 83], "shown": [21, 26, 60, 64, 68, 70, 72, 74, 75, 77], "shrunk": 77, "shuffl": 77, "shut": 2, "shutdown": [0, 54, 64, 65], "si": 5, "sibl": 14, "side": [3, 77], "side_stream_id": 77, "sidestreamidtyp": 77, "sigh": 53, "sigmoid": [14, 77], "signal": 0, "signatur": [7, 47, 77], "signifi": 74, "signific": [3, 5, 21, 53, 71, 72, 74, 75], "significantli": [23, 24, 70, 71, 72, 74, 75, 84, 92], "silu": [14, 77, 78], "similar": [0, 5, 6, 7, 10, 18, 19, 21, 32, 44, 48, 67, 68, 76, 77, 91, 94], "similarli": 10, "simpl": [2, 7, 10, 14, 36, 47, 60, 64, 66, 69, 83, 89], "simpler": 10, "simpleschedul": 94, "simplest": 77, "simpli": [5, 10, 66, 68, 69, 74, 83, 87, 90], "simplic": 17, "simplifi": [5, 17, 68, 74, 77, 88], "simultan": [10, 74], "sin": [0, 77, 78], "sinc": [0, 1, 4, 5, 7, 8, 10, 11, 17, 18, 27, 32, 60, 68, 70, 71, 72, 74, 75, 77, 79, 84, 91, 93, 94], "sinco": 78, "singl": [0, 1, 2, 3, 4, 5, 6, 10, 12, 14, 17, 18, 21, 22, 24, 25, 30, 47, 56, 64, 65, 67, 68, 72, 75, 77, 79, 83, 84, 85, 88, 90, 91, 92, 93], "singleton": [7, 77], "sink": [0, 1, 5, 65, 82], "sink_token_len": 82, "sink_token_length": [5, 65, 82], "sinktokenlength": [0, 1], "sinusoid": 78, "sit": [17, 53], "situaiton": 69, "situat": [10, 53, 59, 70, 74], "size": [0, 1, 2, 5, 6, 8, 9, 10, 11, 18, 20, 21, 23, 24, 25, 26, 32, 47, 50, 51, 52, 59, 65, 67, 68, 69, 70, 71, 72, 73, 75, 77, 78, 79, 82, 87, 88, 92, 94], "size_t": [0, 1], "size_typ": [0, 1], "sizeof": 1, "sizeperhead": [0, 1], "sizetype32": [0, 1], "sizetype64": 1, "skip": [0, 1, 7, 15, 18, 28, 54, 60, 65, 77, 94], "skip_attn": [77, 78], "skip_cross_attn_block": [79, 82], "skip_cross_kv": [78, 82], "skip_encod": 82, "skip_special_token": [65, 88], "skip_tokenizer_init": [32, 65], "skipcrossattnblock": [0, 1], "sku": [70, 72, 74, 75], "skywork": [85, 86, 88], "sleep": 28, "slice": [1, 4, 15, 77, 88], "slice_shap": 15, "sliceinputtyp": 77, "slicen": 1, "slide": [59, 76, 77, 82, 88], "slider": [18, 24, 68], "sliding_window": 79, "sliding_window_caus": 77, "sliding_window_pattern": 79, "slight": [18, 72, 74, 75], "slightli": [0, 2, 9, 26, 72, 75], "slope": [5, 77], "slot": [0, 1, 88], "slot_map": [77, 79], "slotidx": 1, "slotsperpag": 1, "slow": [3, 8, 65, 66, 71], "slower": [17, 71], "slowest": 5, "slurm": [14, 50, 51, 52, 62, 64, 87, 88], "sm": [86, 88], "sm120": 88, "sm80": [86, 88], "sm86": [86, 88], "sm89": [86, 88], "sm90": [86, 88], "small": [5, 8, 10, 14, 23, 24, 70, 72, 74, 75, 77, 84, 87, 88], "smaller": [1, 10, 18, 25, 67, 68, 71, 74, 75, 76, 77, 84, 88], "smallest": [0, 1, 77], "smart": 77, "smaug": [86, 88], "smi": [18, 24, 68, 84], "smile": 53, "smith": [38, 41, 42, 43, 44, 45, 47, 54], "smooth": [17, 65, 88], "smoother": 18, "smoothquant": [7, 23, 59, 88], "smoothquant_v": 65, "snapshot": 68, "snapshot_download": 53, "snip": 68, "snippet": [68, 88, 94], "snshrivas10": 53, "so": [0, 2, 3, 5, 7, 9, 10, 16, 17, 18, 24, 27, 32, 44, 60, 65, 68, 71, 72, 74, 75, 76, 77, 78, 79, 84, 86, 88, 90, 93], "socketst": 0, "softmax": [5, 14, 77, 92], "softplu": 77, "softwar": [3, 5, 14, 59, 66, 88], "solid": 73, "solut": [16, 64, 87, 91], "some": [0, 2, 3, 4, 5, 6, 7, 8, 10, 11, 13, 14, 17, 18, 24, 25, 26, 28, 53, 62, 65, 66, 69, 72, 73, 75, 76, 77, 80, 83, 84, 87, 88, 90, 91, 94], "someth": [14, 32, 46], "sometim": 68, "song": 68, "soon": [0, 19, 20, 21, 22, 23, 32], "sophist": 47, "sora": [30, 56], "sort": [0, 1, 3, 6, 77], "sota": 88, "sourc": [12, 13, 15, 17, 18, 19, 22, 24, 25, 26, 29, 30, 31, 33, 34, 35, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59, 65, 66, 77, 78, 79, 80, 81, 82, 88], "source_root": [50, 51, 52], "sourcetaskvalu": 1, "soyer": [12, 14, 87], "space": [9, 60, 65, 74, 84, 93], "spaces_between_special_token": [65, 88], "span": [17, 24], "spars": [10, 77, 88], "sparsiti": 25, "spatial_norm_dim": 78, "spawn": [36, 45, 61, 62, 64, 70, 83, 87], "spawnprocess": [0, 2], "spec": 25, "spec_decoding_generation_length": [77, 78, 79], "spec_decoding_is_generation_length_vari": [77, 78, 79], "spec_decoding_max_generation_length": [77, 78], "spec_decoding_packed_mask": [77, 78, 79], "spec_decoding_param": [78, 79], "spec_decoding_position_offset": [77, 78, 79], "spec_decoding_us": [77, 78], "specdecconfig": 0, "specdecfastlogitsinfo": 0, "specdecodinggenerationlength": 1, "specdecodinggenerationlengthshost": 1, "specdecodingpackedmask": 1, "specdecodingparam": 78, "specdecodingpositionoffset": 1, "special": [2, 5, 9, 14, 15, 19, 25, 65, 88], "specif": [0, 1, 4, 6, 7, 9, 10, 13, 17, 20, 23, 24, 26, 47, 60, 62, 68, 71, 72, 75, 77, 83, 88, 90, 91], "specifi": [0, 1, 2, 3, 5, 6, 7, 9, 10, 15, 17, 18, 25, 26, 32, 39, 40, 47, 49, 53, 54, 60, 64, 65, 67, 68, 69, 71, 72, 74, 76, 77, 79, 80, 82, 83, 84, 87, 88, 92], "specul": [0, 1, 3, 24, 59, 63, 65, 68, 70, 77, 88], "speculative_config": [18, 24, 39, 48, 49, 65], "speculative_decod": 88, "speculative_decoding_draft_tokens_extern": 79, "speculative_decoding_mod": [25, 65, 68], "speculative_model": [39, 49, 65], "speculativedecod": 0, "speculativedecodingconfig": 0, "speculativedecodingfastlogitsinfo": 0, "speculativedecodingmetr": 0, "speculativedecodingmod": [65, 79, 88], "speculativedecodingmodul": 88, "speculativedecodingoutput": 1, "speed": [14, 20, 24, 25, 68, 69, 75, 88], "speedup": [20, 22, 23, 24], "spent": 0, "split": [1, 4, 5, 9, 14, 68, 71, 72, 77, 84, 88], "split_input_id": 82, "split_prompt_by_imag": 82, "split_siz": 77, "split_size_or_sect": 77, "splittransposecpu": 1, "splittransposecpuinn": 1, "splitwis": 2, "spot": 74, "sq": [23, 85, 88], "sqrt": [5, 77], "squar": [74, 77], "squared_relu": 77, "squeez": [1, 77, 82], "src": [1, 14, 77], "src_seq_len": 77, "srctype": 1, "srun": [14, 26, 50, 51, 52, 62, 87], "sshd": 27, "ssid": 40, "ssm": 77, "ssm_state": 79, "stabil": 24, "stabl": [5, 15, 25, 70, 74, 75, 77, 88], "stack": [15, 24, 60, 77], "stage": [0, 5, 7, 10, 69, 84, 88, 92], "stai": [20, 23, 71, 75], "stand": 14, "standalon": 17, "standard": [10, 14, 16, 19, 69, 77], "starcod": [64, 86, 88], "starcoder1": 85, "starcoder2": [85, 88], "starrickliu": 88, "start": [0, 3, 5, 7, 8, 18, 25, 27, 28, 29, 30, 31, 33, 34, 35, 52, 53, 55, 56, 57, 60, 64, 65, 66, 68, 69, 70, 71, 74, 76, 77, 79, 81, 82, 84, 88], "start_dim": 77, "startup": 87, "stat": [0, 65, 88], "state": [0, 1, 3, 4, 5, 7, 8, 10, 18, 24, 25, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 54, 61, 62, 65, 68, 69, 70, 74, 76, 77, 83, 88, 89, 94], "state_dtyp": 82, "state_or_ptr": 77, "state_s": 82, "statement": 64, "stateptr": 0, "states": 1, "static": [0, 1, 3, 10, 25, 65, 77, 78, 79, 82, 88], "static_batch": [65, 76], "static_cast": 85, "staticbatchingstat": 0, "statist": [0, 3, 10, 26, 65, 68, 88], "statu": 87, "std": [0, 1, 3], "stddev": [26, 34, 35], "stdev": [18, 51, 67, 68, 69, 70], "stdit": 88, "stdout": [18, 51, 67, 68, 69, 70], "steadi": 69, "steady_clock": 0, "step": [0, 1, 5, 6, 7, 8, 10, 13, 14, 16, 17, 19, 24, 28, 47, 59, 61, 62, 65, 66, 68, 69, 70, 77, 82, 87, 91, 92, 93, 94], "still": [5, 15, 17, 18, 24, 66, 68, 70, 72, 77, 82, 84, 88], "stop": [0, 1, 3, 6, 7, 10, 65, 68, 74, 82, 83, 88], "stop_reason": [48, 65, 83, 88], "stop_token_id": [3, 65], "stop_words_data": 82, "stop_words_list": 82, "stopping_criteria": 82, "stoppingcriteria": [82, 88], "stoppingcriterialist": 82, "stoptokenid": [0, 3], "stopword": 0, "stopwordslen": 1, "stopwordslist": 1, "stopwordsptr": 1, "storag": [0, 9, 65], "store": [0, 1, 5, 8, 9, 14, 20, 24, 46, 49, 64, 65, 68, 76, 77, 79, 84, 85, 90, 92, 93], "store_tru": 49, "stored_block": 46, "stori": 53, "str": [13, 17, 42, 43, 65, 77, 78, 79, 82], "strategi": [0, 10, 23, 32, 44, 59, 68, 73, 77, 79, 84, 88], "stream": [0, 1, 2, 3, 14, 25, 26, 32, 34, 35, 36, 37, 47, 65, 67, 77, 82, 84, 87, 88], "stream_ptr": 47, "streaming_llm": 88, "streamingllm": [25, 59, 88], "streamlin": [68, 83], "streamptr": [0, 1, 3], "street": 53, "strenum": [65, 81], "strict": 24, "strict_bound": 77, "strict_dtyp": [77, 78], "stricter": 24, "strictli": 68, "stride": [1, 77, 78], "strike": [10, 46], "string": [0, 1, 3, 13, 40, 65, 68, 77, 82], "string_valu": 8, "string_view": 1, "stringptrmap": 1, "stringvec": 0, "strip": [25, 88], "strip_plan": 25, "strongli": 72, "strongly_typ": [65, 88], "struct": [0, 1], "structur": [0, 4, 7, 10, 47, 65, 77, 84, 88], "structural_tag": 65, "struggl": 53, "student": [38, 41, 42, 44, 45, 47], "studi": [70, 72, 73, 75], "style": [5, 10, 24, 88], "sub": [13, 17, 77], "subclass": [1, 17, 47, 90], "subcommad": 68, "subcommand": [69, 88], "subgraph": [7, 77], "subject": [2, 19, 21, 22, 23, 64, 77, 83, 89], "submiss": 68, "submit": [9, 65, 68], "submit_sync": 65, "submodul": [18, 60, 90], "suboptim": 14, "subscript": 77, "subsequ": [2, 8, 9, 10, 70], "subset": [0, 3, 6, 14, 17, 68, 77], "substanti": [8, 10, 24], "subsystem": 88, "subtract": 7, "succe": [84, 88], "succeed": 82, "success": [3, 20, 24, 69], "successfulli": [10, 28, 72], "sudo": [18, 24, 61, 62, 68], "suffer": 24, "suffici": [71, 72], "suggest": [5, 23, 53, 72], "suit": [5, 68, 69], "sum": [1, 7, 12, 77, 93], "sum_of_token": 77, "summar": [5, 10, 11, 12, 13, 21, 23, 68, 69, 76, 84], "summari": [10, 59], "summat": 77, "sunjiabin17": 88, "super": [7, 12, 15, 17, 86, 87, 90, 94], "superchip": 86, "supplementari": 78, "suppli": [9, 16], "support": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 13, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 32, 40, 47, 50, 51, 52, 53, 59, 62, 63, 65, 69, 70, 72, 74, 75, 76, 77, 78, 80, 83, 87, 88, 89, 90, 91, 92, 93, 94], "supportsinflightbatch": 1, "suppos": 90, "suprem": [38, 41, 42, 44, 45], "sure": [2, 17, 18, 28, 60, 68, 76, 77, 88], "surpass": 5, "surround": [5, 88], "sweep": [14, 20, 74], "sweet": 74, "swept": 21, "swiglu": [25, 77, 88], "switch": [4, 8, 20, 23, 24, 60, 76, 84, 88], "sxm": [20, 25, 70, 72, 73], "sy": 88, "sync": 82, "synchron": [1, 3, 14, 65, 87, 88], "syntax": [77, 83], "synthet": [18, 26, 34, 35, 68, 69], "synthetic_128_128": 68, "synthetic_2048_2048": 70, "synthetic_2048_2048_1000": 70, "system": [8, 14, 18, 20, 26, 29, 30, 40, 50, 51, 52, 55, 56, 59, 60, 62, 69, 71, 83, 86, 88, 89], "systemat": 24, "t": [0, 1, 5, 10, 14, 17, 24, 26, 27, 32, 46, 50, 51, 52, 62, 65, 67, 68, 71, 74, 75, 77, 79, 82, 87], "t5": [5, 6, 85, 86, 88], "tabl": [0, 6, 8, 20, 23, 25, 68, 69, 77, 78, 82, 86, 87, 88], "tactic": 25, "tag": [0, 27, 60, 65], "tailor": [23, 72, 75], "take": [0, 1, 2, 5, 6, 7, 8, 13, 17, 46, 53, 66, 68, 70, 71, 74, 77, 78, 93], "taken": [15, 19, 20, 77], "talk": 53, "tanh": [77, 78], "target": [0, 15, 18, 25, 32, 59, 60, 68, 75, 76, 88], "target_isl": 68, "target_osl": 68, "targetcach": 1, "targetpageid": 1, "targetprob": 1, "targettaskvalu": 1, "tarot": 53, "task": [0, 1, 8, 9, 10, 12, 13, 42, 43, 50, 51, 52, 65, 68, 78, 82, 85, 88, 93], "task_id": [9, 68], "task_vocab_s": 78, "taskid": [0, 1], "taskidtyp": 1, "tasklayermoduleconfig": 1, "tasklayermoduleconfigbind": 1, "tasklayermoduleconfiglistptr": 1, "taskshost": 1, "taskvalu": 1, "taskvalueptr": 1, "taslid": 1, "tayef": 88, "tconstptr": 1, "tcp": 28, "team": [13, 17, 18, 24, 28, 86, 88], "tech": 88, "technic": 59, "techniqu": [5, 7, 10, 14, 19, 24, 66, 71, 72, 73, 76, 85, 88], "technologi": [24, 38, 41, 42, 44, 45, 47], "tekit_2025": 68, "tell": [30, 53, 54, 56, 75, 83], "temb": 78, "temp": 82, "temperatur": [0, 1, 6, 26, 29, 30, 31, 32, 36, 38, 39, 41, 42, 43, 44, 45, 46, 47, 49, 54, 61, 62, 65, 68, 70, 76, 82, 83, 88], "tempfil": [41, 44], "templat": [0, 1, 14, 15], "tempor": 82, "temporari": 2, "ten": [10, 23], "tend": 76, "tensor": [1, 6, 13, 14, 15, 18, 19, 20, 21, 22, 24, 26, 45, 47, 59, 65, 68, 69, 72, 73, 75, 77, 78, 79, 82, 85, 87, 88, 90, 92], "tensor_dict": 82, "tensor_input": 7, "tensor_parallel_s": [45, 46, 49, 50, 51, 52, 65, 70, 71, 72, 75, 76], "tensor_shap": 15, "tensorconstptr": 1, "tensorinfo": 82, "tensorloc": 77, "tensormap": 1, "tensorparallel": [0, 1, 6], "tensorptr": [0, 1], "tensorrt": [1, 3, 5, 6, 7, 11, 12, 19, 22, 24, 25, 26, 29, 30, 31, 32, 33, 34, 35, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 61, 62, 63, 67, 69, 72, 73, 75, 76, 77, 82, 85, 87, 89, 90, 91, 92, 93, 94], "tensorrt_llm": [0, 1, 2, 3, 5, 6, 7, 9, 11, 12, 14, 15, 17, 18, 26, 27, 28, 32, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 60, 61, 62, 65, 68, 69, 70, 72, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 91, 92, 93], "tensorrt_llm_gpt": 14, "tensorrt_llm_rouge1_threshold": 13, "tensorrtllm_backend": [9, 83, 88], "term": [14, 64, 76, 77, 83], "termin": [0, 8, 28, 69, 88], "test": [5, 23, 24, 26, 30, 56, 59, 60, 61, 62, 68, 69, 70, 72, 73, 74, 75, 76, 86, 88, 93], "test_graph_rewrit": 7, "test_trt_llm": [11, 12, 13], "texec": 0, "text": [0, 3, 5, 6, 8, 25, 30, 32, 36, 37, 38, 45, 46, 54, 56, 61, 62, 65, 66, 68, 69, 70, 76, 82, 83, 86, 87, 88, 89], "text_diff": 65, "text_hidden_s": 79, "textattack": 86, "textprompt": 65, "tg_group": 77, "tgt": [14, 77], "tgt_len": [77, 78], "tgt_seq_len": 77, "th": [1, 13, 77], "than": [0, 1, 2, 3, 5, 6, 7, 8, 10, 14, 18, 19, 20, 21, 23, 24, 25, 60, 65, 66, 68, 69, 70, 71, 72, 74, 76, 77, 82, 84, 87, 88, 92], "thank": 88, "thecodewrangl": 88, "thei": [0, 1, 3, 5, 6, 9, 14, 15, 17, 24, 48, 60, 65, 68, 70, 72, 74, 75, 76, 77, 79, 85, 88], "them": [0, 3, 4, 7, 10, 11, 18, 24, 50, 51, 52, 65, 66, 67, 68, 71, 73, 74, 76, 77, 82, 84, 90], "theoret": 84, "theori": 76, "therebi": [2, 76], "therefor": [11, 17, 69, 77, 87, 93], "thermal": 68, "theta": 77, "thi": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21, 23, 24, 25, 26, 27, 28, 32, 36, 40, 47, 49, 50, 51, 52, 53, 60, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 82, 83, 84, 85, 87, 88, 89, 90, 91, 92, 93, 94], "thin": 17, "thing": [6, 28, 38, 41, 42, 44, 45, 47, 74, 75], "think": [24, 46, 73], "third": [3, 88], "those": [3, 5, 6, 13, 14, 16, 18, 24, 25, 26, 67, 69, 70, 75, 77, 78, 85], "though": [17, 74, 84], "thread": [0, 1, 5, 32, 64, 68, 82], "three": [2, 3, 13, 23, 24, 76, 77, 85, 90, 91, 92], "threshold": [0, 24, 77, 82], "throttl": 68, "through": [0, 5, 6, 7, 10, 14, 15, 16, 18, 24, 25, 26, 60, 66, 68, 70, 71, 72, 74, 75, 78, 83, 88], "throughout": [70, 73], "throughput": [0, 3, 5, 19, 20, 21, 51, 59, 67, 72, 74, 75, 76, 88, 92], "throw": [0, 1], "thu": [8, 17, 18, 24, 60, 77, 84], "thumb": [5, 71, 87], "ti": 5, "tiiuae": 68, "time": [0, 1, 2, 3, 5, 8, 9, 10, 11, 14, 18, 21, 23, 24, 25, 38, 41, 42, 43, 44, 45, 53, 59, 60, 65, 66, 67, 68, 69, 70, 72, 73, 74, 76, 77, 82, 87, 88, 93], "time_embed_dim": 78, "time_encod": 82, "time_point": 0, "timedelta": 65, "timedout": 0, "timelin": 13, "timeout": [0, 26, 32, 65, 88], "timepoint": 0, "timestamp": 0, "timestep": [78, 79], "timestepembed": 78, "timingmetr": 0, "tini": 53, "tinyllama": [26, 29, 31, 34, 36, 38, 40, 41, 42, 43, 44, 45, 46, 47, 48, 53, 54, 55, 57, 61, 62, 64, 83, 89], "tip": 59, "titl": 40, "tle": 11, "tllm_checkpoint_16gpu_tp8_pp2": 71, "tllm_ckpt_dir": 12, "tllm_engine_dir": 12, "tllm_kei": [15, 78], "tllm_llmapi_build_cach": 88, "tllm_llmapi_enable_nvtx": 67, "tllm_log_level": 87, "tllm_nvtx_debug": 67, "tllm_override_layer_num": 88, "tllm_profile_record_gc": 67, "tllm_profile_start_stop": 67, "tllm_to_externel_key_dict": 15, "tllm_torch_profile_trac": 67, "tllm_trace_model_forward": 88, "tllm_weight": 15, "tllmruntim": [1, 6, 87], "tlntin": 88, "tmp": [9, 11, 51, 67, 68, 71], "tmp9so41y3r": 68, "tmpowsrb_f4": 68, "tmpxhdvasex": 68, "to_arrai": 77, "to_dict": [65, 79], "to_json_fil": 79, "to_layer_quant_config": 79, "to_legacy_set": 80, "to_str": [0, 1, 3], "to_trt": 79, "tobyt": 1, "todo": [1, 49, 77], "togeth": [3, 5, 6, 9, 14, 16, 19, 24, 25, 82, 85, 88], "toggl": 67, "toi": 74, "toitensor": 0, "tojsonstr": 0, "tok": [19, 21, 22, 75], "token": [0, 1, 2, 3, 4, 5, 6, 8, 10, 14, 18, 19, 22, 23, 24, 25, 26, 27, 34, 35, 40, 46, 47, 51, 59, 65, 67, 68, 69, 70, 72, 73, 75, 77, 78, 79, 82, 83, 84, 85, 88, 90, 91, 92], "token_drop": 78, "token_end": 65, "token_extra_id": 46, "token_id": [32, 46, 47, 48, 65], "token_ids_diff": 65, "token_range_retention_config": 65, "token_start": 65, "token_type_id": [79, 82], "tokenend": 0, "tokenextraid": 1, "tokenextraidtyp": 1, "tokenid": 1, "tokenidtyp": [0, 1], "tokenization_utils_bas": 65, "tokenizer_dir": [12, 14, 83, 87], "tokenizer_image_token": 82, "tokenizer_max_seq_length": [65, 72, 79, 81], "tokenizer_mod": 65, "tokenizer_revis": 65, "tokenizer_str": [0, 3], "tokenizerbas": 65, "tokenizerstr": [0, 3], "tokenlogprob": 65, "tokenrangeretentionconfig": [0, 65], "tokenrangeretentionprior": 0, "tokens_per_block": [8, 25, 82, 88, 93], "tokensperblock": [0, 1, 6], "tokensperstep": 1, "tokensprompt": 65, "tokenstart": 0, "tokyo": [30, 56], "toler": 23, "tomodulenam": 1, "tomoduletyp": 1, "tonylek": 88, "too": [3, 5, 18, 70, 74, 87], "took": 70, "tool": [2, 13, 18, 59, 64, 68, 88], "tool_cal": 83, "toolkit": [16, 17, 23, 24, 62, 91], "top": [0, 5, 6, 10, 14, 16, 65, 77, 88], "top1": 24, "top_k": [6, 65, 82, 88], "top_p": [6, 36, 38, 39, 41, 42, 43, 44, 45, 46, 47, 49, 54, 61, 62, 65, 70, 76, 82, 83], "top_p_decai": [65, 82], "top_p_min": [65, 82], "top_p_reset_id": [65, 82], "topenkoff": 88, "topic": 75, "topk": [0, 1, 4, 6, 10, 24, 77, 88], "topk_logit": 3, "topklastdim": 77, "topklogit": 3, "topkmedusahead": 1, "topktopp": [0, 6], "topmodelmixin": [17, 79], "topn": 24, "topp": [0, 1, 6, 88], "toppdecai": [0, 1, 6], "toppmin": [0, 1, 6, 65], "toppresetid": [0, 1, 6], "torch": [5, 15, 47, 54, 60, 61, 62, 65, 68, 77, 82, 87, 90], "torchaudio": [61, 62], "torchvis": [61, 62], "tostr": [0, 1], "total": [0, 1, 4, 5, 6, 10, 13, 15, 18, 25, 26, 68, 69, 70, 71, 84, 93], "total_lat": [19, 22], "total_token": 83, "totalaccepteddrafttoken": 0, "totaldrafttoken": 0, "totalgentoken": 1, "totalnumpag": 1, "totensor": 0, "touch": [27, 90], "tp": [0, 2, 4, 6, 9, 14, 18, 19, 20, 21, 22, 23, 24, 26, 51, 68, 69, 70, 77, 88], "tp1": [19, 20, 21], "tp2": 68, "tp4": 24, "tp4ep2": 24, "tp8": [21, 24], "tp8ep2": 24, "tp_1_pp_1": 68, "tp_dim": [15, 78], "tp_group": [77, 78], "tp_rank": [15, 77, 78], "tp_size": [4, 9, 13, 14, 15, 17, 26, 33, 50, 52, 68, 69, 71, 77, 78, 81, 88], "tp_split_dim": 78, "tpot": [22, 69], "tprank": 1, "tpsize": 1, "tqdm": [15, 65, 88], "trace": [17, 67, 87], "track": [5, 65, 77], "trade": 8, "tradeoff": [23, 24, 72], "tradit": 0, "train": [10, 12, 13, 14, 16, 17, 20, 23, 68, 77, 87, 90], "trait": 88, "transa": 77, "transb": 77, "transceiv": [0, 65], "transfer": [0, 2, 14, 47, 65, 88], "transform": [0, 4, 5, 10, 12, 13, 14, 15, 25, 26, 32, 65, 79, 83, 84, 86, 87, 88, 90, 91, 93], "translat": [76, 88], "transmiss": 2, "transmit": 2, "transpos": [1, 13, 77], "transposit": 77, "travers": 14, "treat": [5, 24, 77], "tree": [0, 68, 82, 87, 93], "tri": 94, "tricki": 79, "trigger": [5, 7, 14, 25, 32, 54, 64], "trim": 1, "trimpool": 1, "triton": [8, 9, 10, 14, 16, 59, 66, 88], "tritonserv": 88, "trivial": 14, "troubleshoot": [59, 88], "trt": [0, 2, 3, 4, 5, 6, 7, 8, 9, 14, 15, 20, 27, 41, 44, 68, 74, 77, 79, 81, 82, 84, 87, 88, 92], "trt_ckpt": [9, 11, 13, 87], "trt_engin": [9, 11, 13, 87], "trt_root": 18, "trt_tensor": [14, 77], "trtdatatyp": 1, "trtgptmodel": 84, "trtgptmodeloptionalparam": 88, "trtgptmodelv1": 88, "trtllm": [8, 9, 11, 12, 13, 14, 17, 18, 29, 30, 31, 32, 33, 34, 35, 36, 37, 50, 55, 56, 57, 59, 64, 65, 68, 69, 72, 73, 74, 75, 84, 87, 88], "trtllm_dg_jit_use_nvcc": 18, "trtllm_disable_kv_cache_transfer_overlap": 2, "trtllm_disable_unified_convert": 15, "trtllm_enable_kvcache_receive_parallel": 2, "trtllm_enable_mmha_multi_block_debug": 68, "trtllm_enable_pdl": [18, 24, 68], "trtllm_force_xqa": 5, "trtllm_kvcache_send_max_concurrency_num": 2, "trtllm_kvcache_transfer_buffer_s": 2, "trtllm_kvcache_transfer_use_async_buff": 2, "trtllm_mmha_blocks_per_sequ": 68, "trtllm_mmha_kernel_block_s": 68, "trtllm_model": 15, "trtllm_modules_to_hf_modul": 82, "trtllm_parallel_cache_send": 2, "trtllm_pdl_overlap_ratio": 68, "trtllm_precompiled_loc": 60, "trtllm_prefetch_ratio": 68, "trtllm_request_kv_cache_concurr": 2, "trtllm_serv": 26, "trtllm_try_zcopy_for_kvcache_transf": 2, "trtllm_use_mpi_kvcach": 2, "trtllm_use_precompil": 60, "trtllm_use_ucx_kvcach": 2, "trtllmattent": 92, "trtlmmdatatyp": 0, "true": [0, 1, 3, 6, 7, 8, 10, 13, 18, 24, 26, 32, 38, 39, 43, 44, 46, 47, 48, 49, 51, 53, 65, 67, 68, 69, 72, 75, 77, 78, 79, 80, 82, 84, 87, 88], "true_output_valu": 77, "true_valu": 77, "truncat": [65, 88], "truncate_prompt_token": [65, 88], "trust": 65, "trust_remote_cod": [26, 65, 88], "try": [0, 1, 3, 12, 17, 48, 53, 64, 69, 72, 74, 75, 76, 83, 84, 87, 89], "tsuji": 68, "ttensor": 1, "ttft": [69, 72, 74, 75, 76, 88], "ttim": 88, "ttl": 24, "tunabl": 73, "tune": [0, 2, 3, 10, 20, 23, 24, 25, 59, 65, 68, 69, 72, 75, 78, 79, 82, 83, 84, 88], "tuner": 0, "tupl": [0, 1, 77, 78, 82, 94], "turn": [5, 6, 8, 10, 60, 72, 82, 84, 88], "tushar": 88, "tweak": 76, "twice": 14, "two": [0, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 17, 20, 24, 25, 26, 30, 56, 60, 64, 68, 70, 72, 74, 76, 77, 78, 80, 88, 91, 93, 94], "twofold": 10, "twoshot": 77, "txt": [17, 18, 51, 62, 67, 68, 70, 83, 88], "type": [1, 2, 3, 5, 6, 7, 9, 13, 14, 20, 23, 25, 26, 29, 30, 31, 34, 35, 40, 46, 47, 49, 56, 65, 68, 72, 75, 77, 79, 81, 82, 83, 85, 86, 87, 88, 90, 91, 92, 93], "typedef": [0, 1], "typenam": [0, 1, 14], "typetrait": 0, "typic": [0, 2, 7, 12, 14, 17, 23, 26, 62, 64, 71, 72, 75, 76, 80, 82, 84, 88, 90], "typo": 88, "u": [1, 7, 27, 38, 41, 42, 43, 44, 45, 54, 68, 69, 88], "ub": 77, "ub_oneshot": 68, "ub_tp_siz": 68, "ubuntu": [61, 62, 88, 89], "uc_handl": 1, "uc_ptr": 1, "uc_va": 1, "ucx": [2, 88], "ucx_cuda_copy_async_mem_typ": 2, "ucx_cuda_copy_dmabuf": 2, "ucx_info": 2, "ucx_memtype_cach": 2, "ucx_rndv_frag_mem_typ": 2, "ucx_rndv_pipeline_error_handl": 2, "uid": [0, 82], "uint16_t": 0, "uint32": 1, "uint32_t": [0, 1, 77], "uint64": [1, 8], "uint64_t": [0, 1], "uint8": 1, "uint8_t": [0, 1], "uintptr_t": 1, "uk_bgemm": 24, "ulimit": [60, 87], "ultim": 71, "ulyss": 88, "unabl": [62, 74], "unaccept": 72, "unari": 77, "unaryoper": 77, "unbind": 77, "uncas": 86, "uncertainti": 10, "unchang": [10, 75, 77], "uncommon": 14, "undefin": 77, "under": [0, 23, 25, 60, 64, 68, 69, 87, 88], "underli": [0, 1, 7, 10], "underlying_type_t": 1, "underlyingtyp": [0, 1], "underscor": 72, "understand": [59, 60, 67], "understood": 74, "underutil": 10, "uneven": 88, "unevenli": 24, "unexpect": [87, 88], "unfinish": 0, "unfus": 77, "unfuse_qkv_project": 79, "ungath": 1, "unguid": 40, "unif": 88, "unifi": [13, 17, 23, 88], "uniform": [68, 69, 77], "uniniti": 92, "uninstal": 62, "union": [65, 77], "uniqu": [0, 5, 6, 9, 10, 13, 25, 65, 68], "unique_ptr": [0, 1], "unique_token": 46, "uniqueconstptr": 1, "uniqueptr": 1, "uniquetoken": 1, "unit": [1, 15, 36, 38, 39, 41, 42, 43, 44, 45, 47, 49, 54, 59, 60, 61, 62, 68, 70, 76, 83, 89], "univers": [38, 41, 42, 44, 45, 47], "unless": [0, 32, 65, 71, 75, 76], "unlik": [8, 10], "unlock": 66, "unnecessari": [7, 88, 90, 94], "unneed": [5, 24], "unordered_map": [0, 1, 3], "unpatchifi": 79, "unschedul": 74, "unset": 76, "unsign": 1, "unspecifi": [25, 26, 77], "unsqueez": [1, 77], "unstabl": 17, "unsupport": 88, "until": [0, 1, 3, 6, 8, 10], "untouch": 77, "unus": [0, 68], "up": [0, 5, 6, 9, 10, 18, 20, 21, 24, 25, 40, 68, 74, 75, 88, 93], "up_proj": 15, "upcast": 77, "upcast_attent": 78, "upcast_softmax": 78, "upcom": [23, 93], "updat": [0, 10, 14, 15, 17, 18, 21, 25, 27, 47, 60, 65, 77, 82, 87, 93], "update_from_dict": 65, "update_key_map": 15, "update_kv_cache_typ": 65, "update_output_ids_by_offset": 82, "update_resourc": [91, 93], "update_strategi": 77, "updatenumreturnbeam": 0, "updatespositionid": 1, "upgrad": [61, 62, 83], "uplift": [72, 74, 75], "upon": [10, 69, 75, 87, 88], "upper": [68, 77, 84], "uq_qr_gemm": 24, "url": [26, 30, 34, 35, 56, 60, 61, 62, 88], "us": [0, 1, 2, 3, 4, 5, 6, 8, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 32, 36, 37, 40, 43, 50, 51, 52, 53, 59, 60, 61, 62, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 77, 78, 79, 80, 82, 83, 85, 87, 88, 89, 90, 91, 92, 93, 94], "usabl": 89, "usag": [0, 5, 7, 14, 17, 19, 22, 25, 26, 36, 59, 65, 68, 75, 76, 77, 83, 88, 92], "use_beam_hyp": 82, "use_beam_search": [44, 65, 88], "use_cach": [77, 78, 79], "use_context_fmha_for_gener": 88, "use_cuda_graph": [18, 51, 69], "use_custom_all_reduc": 88, "use_diff_of_squar": 77, "use_dynamic_tre": [39, 65], "use_embedding_shar": 88, "use_fp32_acc": 77, "use_fp8": 78, "use_fp8_context_fmha": [5, 25, 68, 88], "use_fused_mlp": [25, 68, 88], "use_gemm_allreduce_plugin": 82, "use_gpt_attention_plugin": 82, "use_gpu_direct_storag": 82, "use_implicit_relative_attent": 78, "use_kv_cach": [78, 82], "use_logn_sc": 78, "use_lora": 79, "use_lora_plugin": 82, "use_mamba_conv1d_plugin": 82, "use_meta_recip": 65, "use_modelopt_ckpt": 49, "use_modelopt_quant": 17, "use_mrop": 65, "use_one_more_block": 82, "use_paged_context_fmha": [5, 8, 25, 68, 72, 75], "use_parallel_embed": [13, 14, 79], "use_preload": 79, "use_prompt_tun": [79, 88], "use_py_sess": 87, "use_refit": 65, "use_relaxed_acceptance_for_think": [24, 65], "use_runtime_default": 82, "use_safetensors_load": 79, "use_strip_plan": 65, "use_tqdm": 65, "use_variable_beam_width_search": 82, "usebantoken": 0, "usebanword": 0, "usecrossattent": 1, "usedefaultvalu": 1, "usednumblock": 0, "usedraftlogit": 1, "usedraftlogitshost": 1, "usedynamictre": 0, "usedynamictreehost": 1, "useexpliciteosstop": 0, "usefrequencypenalti": 0, "usegemmallreduceplugin": 1, "usegptattentionplugin": [1, 6], "usegpudirectstorag": 0, "uselanguageadapt": 1, "useloraplugin": 1, "usemambaconv1dplugin": 1, "usemaxlengthstop": 0, "useminlen": 0, "useminlength": 0, "useminp": 0, "usemrop": 1, "usenorepeatngrams": 0, "useoccurrencepenalti": 0, "usepackedinput": 1, "usepagedst": 1, "usepenalti": 0, "usepositionembed": 1, "usepresencepenalti": 0, "useprompttun": 1, "user": [0, 2, 3, 5, 6, 7, 8, 9, 14, 15, 16, 17, 18, 22, 23, 24, 26, 27, 29, 30, 39, 40, 44, 47, 48, 49, 55, 56, 60, 64, 65, 67, 68, 69, 74, 75, 76, 77, 79, 83, 84, 85, 87, 88], "user_buff": [25, 72], "userandomacceptancethreshold": 1, "userbuff": 88, "userepetitionpenalti": 0, "userwarn": 62, "useshapeinfer": 1, "usespecdecod": 1, "usestopword": 0, "usetemp": 0, "usetemperatur": 0, "usetokentypeembed": 1, "usevariablebeamwidthsearch": 0, "usr": [13, 18, 26, 29, 30, 31, 33, 34, 35, 62, 68], "usual": [14, 17, 62, 69, 70, 75, 77, 93], "util": [0, 1, 2, 5, 6, 10, 14, 18, 19, 24, 25, 36, 62, 66, 67, 68, 72, 75, 76, 84, 88, 92], "uv_gemm": 24, "uvm": [0, 1], "v": [1, 2, 5, 6, 9, 18, 19, 20, 23, 24, 59, 65, 77, 79, 82, 85, 86, 87, 90, 92], "v0": [9, 19, 20, 21, 22, 66, 68, 69, 86, 88], "v1": [26, 29, 30, 31, 34, 36, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 53, 54, 55, 56, 57, 61, 62, 64, 83, 86, 88, 89], "v10": 88, "v100": 88, "v12": 88, "v2": [23, 85, 88], "v3": [26, 67, 85, 86, 88], "v9": 21, "v_dim": 77, "v_head_dim": [77, 78], "v_proj": [15, 90], "vacat": [38, 41, 42, 44, 45], "valid": [0, 1, 3, 10, 65, 69, 77, 82], "validate_positive_valu": 65, "validatevec": 1, "validationerror": 65, "validmpiconfig": 1, "valu": [0, 1, 2, 5, 6, 8, 9, 11, 13, 14, 15, 18, 19, 20, 25, 26, 32, 54, 65, 68, 70, 72, 74, 76, 77, 79, 80, 81, 82, 84, 85, 87, 88, 92, 93, 94], "valuabl": 24, "value_typ": 0, "valuestatu": 1, "vanilla": [5, 92], "vanillaattent": 92, "var": 77, "vari": [21, 74, 75, 93], "variabl": [0, 1, 6, 15, 18, 21, 24, 50, 51, 52, 59, 62, 65, 67, 68, 87, 88], "variabledraftlength": 1, "varianc": [72, 74, 75, 77], "variant": [0, 3, 5, 17, 19, 64, 77, 83, 88, 92], "varieti": [68, 70, 88], "variou": [5, 10, 16, 68, 72, 74, 88], "varnam": 1, "vartyp": 1, "vboost": [18, 24, 68], "vbw": 88, "ve": [24, 53], "vec": 1, "vec2": 77, "veclogprob": 0, "vectoken": 0, "vectokenextraid": [0, 1], "vector": [0, 1, 3, 5, 6, 9, 77], "vecuniquetoken": [0, 1], "verbatim": 79, "verbos": [25, 26, 68], "veri": [5, 13, 14, 16, 23, 70, 71, 72, 88], "verif": [0, 10, 65], "verifi": [10, 59, 75, 77, 88], "verificationsets": 0, "versa": 8, "version": [0, 1, 2, 5, 6, 13, 15, 17, 18, 24, 26, 32, 60, 62, 68, 70, 77, 83, 87, 88, 89], "vertic": 77, "vertical_strid": 78, "via": [0, 2, 10, 24, 50, 51, 52, 53, 60, 62, 68, 72, 73, 75, 76, 77, 88, 89], "vice": [8, 54], "vicuna": [10, 39, 49], "video": [30, 56, 68, 82, 86, 88], "video_grid_thw": 82, "video_path": 82, "video_preprocess": 82, "video_url": [30, 56], "view": [1, 77, 82], "vila": [30, 56, 85, 86, 88], "vinyl": 68, "violat": 88, "virtual": [0, 1, 78], "vision": [82, 85, 86, 88], "vision_grid_thw": 82, "vision_length": 77, "vision_model_typ": 79, "vision_start": 77, "vision_token_mask": 78, "visit": [10, 24, 88], "visual": [74, 88], "visual_engine_dir": 82, "visual_featur": 82, "visualize_network": [25, 65, 88], "vit": 88, "vital": [7, 23], "vl": [26, 30, 35, 56, 68, 86, 88], "vlm": [86, 88], "vocab": [77, 82], "vocab_embed": [12, 15], "vocab_s": [0, 13, 15, 65, 78, 79, 82, 90], "vocab_size_pad": 82, "vocabs": [1, 6], "vocabsizepad": [0, 1], "vocabulari": [0, 1, 6, 8, 10, 69, 78, 82], "void": [0, 1, 3, 14], "volta": 88, "volum": [1, 60, 68], "volumenonneg": 1, "vonjackustc": 88, "vote": [38, 41, 42, 44, 45], "vulner": 88, "vultureprim": 88, "w": [1, 22, 24, 26, 77, 79, 85, 86, 88], "w1": 77, "w4a": [85, 88], "w4a16": [13, 23, 59, 65, 79], "w4a16_awq": [13, 17, 32, 54, 65], "w4a16_gptq": [13, 65], "w4a8": [23, 88], "w4a8_awq": [13, 17, 65], "w4a8_qserve_per_channel": 65, "w4a8_qserve_per_group": 65, "w4aint8": 88, "w8a": 85, "w8a16": [13, 23, 59, 65, 79], "w8a16_gptq": 65, "w8a8": [20, 23, 59], "w8a8_sq_per_channel": [13, 65], "w8a8_sq_per_channel_per_tensor_plugin": [65, 79], "w8a8_sq_per_channel_per_token_plugin": [65, 79], "w8a8_sq_per_tensor_per_token_plugin": [65, 79], "w8a8_sq_per_tensor_plugin": [65, 79], "wa": [0, 1, 3, 5, 6, 13, 62, 64, 68, 69, 70, 72, 74, 75, 76, 78, 85, 87, 88, 90, 94], "wai": [2, 5, 7, 16, 24, 45, 47, 64, 66, 68, 70, 72, 77, 84, 88], "wait": [0, 1, 3, 17, 32, 65, 66, 68, 77], "walk": [30, 53, 56, 70, 71, 72], "wang1120": 88, "wangkuiyi": 88, "want": [5, 10, 17, 24, 28, 62, 67, 68, 72, 74, 76, 77, 87, 88, 90], "warm": 93, "warmup": [18, 67, 68, 70, 88, 92, 93], "warn": [5, 25, 26, 68, 69, 84], "warp": 88, "watch": 75, "wdkv": 24, "wdq": 24, "we": [1, 2, 4, 6, 7, 9, 10, 11, 13, 17, 18, 22, 23, 24, 26, 27, 28, 38, 41, 42, 44, 45, 53, 54, 60, 62, 64, 67, 68, 69, 70, 71, 72, 74, 75, 77, 82, 83, 87, 88, 90], "weapon": 46, "wear": 46, "web": [16, 28], "weig": 77, "weight": [0, 1, 4, 9, 17, 19, 20, 23, 24, 25, 26, 45, 59, 65, 66, 69, 70, 71, 72, 77, 78, 79, 82, 83, 88], "weight_index": 77, "weight_load": 78, "weight_only_groupwise_quant_matmul": 85, "weight_only_precis": 88, "weight_spars": [25, 65], "weight_stream": [11, 25, 65], "weightonlygroupwisequantmatmulplugin": 85, "weights_dict": 17, "weights_scaling_factor": [13, 15], "weightsinpoint": 1, "weightsoutpoint": 1, "well": [5, 6, 14, 16, 20, 32, 67, 74, 75, 85, 86], "were": [0, 1, 10, 13, 17, 19, 23, 69, 71, 74, 88], "weren": 62, "wget": 87, "what": [2, 3, 30, 53, 56, 59, 60, 67, 68, 70, 72, 74, 75], "whatev": 1, "wheel": [60, 62, 88], "when": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 14, 15, 17, 18, 22, 23, 25, 27, 32, 47, 59, 60, 62, 65, 67, 68, 70, 72, 74, 75, 76, 77, 78, 79, 82, 83, 84, 85, 87, 88, 90, 92, 93], "whenev": 1, "where": [0, 1, 2, 5, 6, 8, 10, 13, 14, 19, 23, 24, 26, 29, 31, 32, 53, 55, 57, 65, 68, 69, 72, 74, 76, 77, 82, 83, 85, 88, 94], "wherea": [0, 13, 74], "whether": [0, 1, 2, 3, 5, 9, 25, 65, 71, 72, 75, 77, 78, 82, 91, 92], "which": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 17, 19, 23, 24, 25, 26, 60, 62, 64, 65, 67, 68, 70, 72, 74, 75, 76, 77, 79, 80, 82, 83, 84, 85, 88, 89, 91, 92, 94], "while": [0, 1, 4, 7, 8, 10, 14, 17, 19, 20, 22, 23, 24, 62, 66, 68, 70, 71, 72, 73, 74, 75, 76, 77, 84, 85, 88, 92], "whisper": [85, 86, 88], "whisperencod": 79, "whl": [18, 60, 61, 62], "who": 64, "whole": [1, 65, 66, 77], "whose": [2, 8, 13, 24, 78], "why": [0, 2, 14, 65, 72, 74, 75, 77, 84], "wide": [0, 4, 70], "width": [0, 1, 5, 6, 35, 65, 78, 82, 84, 88], "window": [0, 1, 10, 25, 59, 65, 68, 77, 82, 88], "window_s": 5, "windows": 0, "wip": 24, "wireless": 40, "wirelessaccesspoint": 40, "wise": [7, 65, 77, 88], "wish": 8, "wit": 46, "with_ssh": 27, "within": [1, 2, 5, 10, 14, 46, 65, 68, 71, 72, 74, 75, 77, 83, 93], "without": [0, 1, 3, 5, 10, 14, 15, 18, 23, 24, 25, 32, 46, 66, 68, 72, 75, 77, 79, 88, 90, 92], "wkr": 24, "wo": [15, 24, 88], "wo_gemm": 24, "won": [62, 71], "word": [0, 3, 5, 65, 77, 82, 88], "word_dict": 82, "word_embed": 15, "word_embeddings_layernorm": 15, "work": [5, 6, 7, 10, 14, 17, 18, 32, 47, 50, 51, 52, 54, 60, 62, 66, 69, 73, 77, 82, 85, 87, 88, 90], "workaround": [15, 18, 88], "workdir": [26, 50, 51, 52, 60], "worker": [14, 25, 26, 65, 68, 84, 88], "workerexecutablepath": 0, "workflow": [5, 6, 12, 13, 18, 32, 59, 64, 69, 70, 72, 73, 77, 83, 87, 88, 89], "workload": [4, 14, 25, 67, 68, 70, 72, 73, 74, 75], "workspac": [1, 25, 26, 65, 68, 77, 84, 88], "workstat": 20, "world": [0, 2, 7, 18, 25, 50, 51, 52, 65, 66, 68, 70, 71, 72, 77], "world_config": 82, "world_siz": [13, 17, 77, 88], "worldconfig": [0, 6, 82], "worldsiz": 1, "wors": [10, 25, 72], "worst": [74, 75], "worth": [5, 72, 75], "would": [0, 7, 10, 68, 70, 72, 74, 76, 77, 90], "wouldn": 46, "wpa2": 40, "wqr": 24, "wrap": [0, 1, 14, 25, 64, 70, 77, 80, 82, 88], "wrapper": [1, 7, 17, 92], "write": [1, 8, 15, 24, 25, 59, 77, 87], "written": [14, 68, 77], "wrong": [10, 46, 88], "wsl": 88, "wuk": 24, "wuq": 24, "wuv": 24, "www": 88, "x": [0, 1, 3, 6, 9, 11, 26, 68, 77, 78, 79, 83, 85, 88], "x86": 8, "x86_64": 86, "xcomposer2": 88, "xgrammar": [0, 3, 40, 88], "xl": 88, "xml": 3, "xor": 77, "xqa": 88, "xxx": [15, 17, 87], "xxx_plugin": 80, "xy": 77, "y": [2, 3, 18, 22, 27, 60, 61, 62, 68, 77, 79, 85], "y_bia": 77, "yaml": [26, 68, 69], "yarn": 77, "ye": [2, 77, 84], "yeah": 53, "yelp": 86, "yen": 68, "yet": [0, 6, 17, 20, 24, 77, 94], "yield": [8, 32, 72, 74], "yiyixu": [30, 56], "yml": [18, 26, 33, 68, 69], "york": [26, 29, 31, 55, 57, 83], "you": [3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 16, 17, 18, 23, 24, 25, 26, 27, 28, 29, 30, 32, 40, 41, 44, 47, 50, 51, 52, 53, 54, 55, 56, 59, 60, 62, 64, 65, 68, 69, 71, 72, 73, 74, 75, 76, 77, 82, 83, 84, 87, 88, 89, 90, 92], "your": [8, 9, 10, 16, 17, 18, 23, 25, 27, 28, 32, 53, 60, 62, 64, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 83, 87, 90, 92, 93], "your_data_path": 18, "your_dockerhub_usernam": [27, 28], "your_model_path": 18, "your_public_kei": 28, "your_work_path": 18, "yourself": 89, "yuhuili": 39, "yyi": 87, "z": 77, "zars19": 88, "zero": [0, 1, 3, 15, 64, 65, 77, 78, 85, 87], "zero_is_placehold": 77, "zip": 47, "zjli2013": 88, "zoo": 88}, "titles": ["Executor", "Runtime", "Disaggregated-Service (experimental)", "Executor API", "Expert Parallelism in TensorRT-LLM", "Multi-Head, Multi-Query, and Group-Query Attention", "C++ GPT Runtime", "Graph Rewriting Module", "KV cache reuse", "Run gpt-2b + LoRA using Executor / cpp runtime", "Speculative Sampling", "Running With Weight Streaming to Reduce GPU Memory Consumption", "Adding a Model", "TensorRT-LLM Checkpoint", "Model Definition", "TensorRT-LLM Model Weights Loader", "TensorRT-LLM Architecture", "TensorRT-LLM Build Workflow", "How to get best performance on DeepSeek-R1 in TensorRT-LLM", "Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100", "H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token", "H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLM", "New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget", "Speed up inference with SOTA quantization techniques in TRT-LLM", "Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs", "trtllm-build", "trtllm-serve", "Build the TensorRT-LLM Docker Image", "Develop TensorRT-LLM on Runpod", "Curl Chat Client", "Curl Chat Client For Multimodal", "Curl Completion Client", "LLM Common Customizations", "Deepseek R1 Reasoning Parser", "Genai Perf Client", "Genai Perf Client For Multimodal", "LLM Examples Introduction", "LLM Examples", "Automatic Parallelism with LLM", "Generate Text Using Eagle Decoding", "Generate text with guided decoding", "Generate text", "Generate Text Asynchronously", "Generate Text in Streaming", "Generate text with customization", "Distributed LLM Generation", "Get KV Cache Events", "Control generated text using logits processor", "Generate Text Using Lookahead Decoding", "Generate Text Using Medusa Decoding", "Llm Mgmn Llm Distributed", "Llm Mgmn Trtllm Bench", "Llm Mgmn Trtllm Serve", "Generate text with multiple LoRA adapters", "Generation with Quantization", "OpenAI Chat Client", "OpenAI Chat Client", "OpenAI Completion Client", "Online Serving Examples", "Welcome to TensorRT-LLM\u2019s Documentation!", "Building from Source Code on Linux", "Installing on Grace Hopper", "Installing on Linux", "Key Features", "API Introduction", "API Reference", "Overview", "Performance Analysis", "TensorRT-LLM Benchmarking", "Overview", "Benchmarking Default Performance", "Deciding Model Sharding Strategy", "FP8 Quantization", "Performance Tuning Guide", "Tuning Max Batch Size and Max Num Tokens", "Useful Build-Time Flags", "Useful Runtime Options", "Functionals", "Layers", "Models", "Plugin", "Quantization", "Runtime", "Quick Start Guide", "Memory Usage of TensorRT-LLM", "Numerical Precision", "Support Matrix", "Troubleshooting", "Release Notes", "PyTorch Backend", "Adding a New Model in PyTorch Backend", "Architecture Ovewiew", "Attention", "KV Cache Manager", "Scheduler"], "titleterms": {"": [5, 20, 23, 59], "0": 88, "000": [20, 21], "1": [12, 14, 18, 60, 69, 84, 88], "10": [20, 88], "100m": 20, "11": 88, "12": [21, 88], "13": 88, "13b": 21, "14": 88, "15": 88, "16": 88, "17": 88, "18": 88, "180b": 19, "19": 88, "2": [12, 18, 22, 60, 84, 88], "2b": 9, "3": [12, 14, 18, 68, 69, 84, 86], "4": [12, 18, 20], "405b": [14, 69], "4x": 22, "5": 18, "6": [18, 19], "6x": 20, "7": 88, "70b": [14, 19, 22, 68, 69], "7x": 19, "8": 88, "8b": 69, "9": 88, "As": 3, "For": [30, 35], "In": [3, 5, 66], "Not": 84, "One": [24, 60], "The": [3, 85], "To": 70, "With": [11, 66], "a100": [19, 20], "about": [10, 26, 66, 71], "accept": 24, "access": 27, "account": 28, "accuraci": 23, "achiev": [20, 21], "acknowledg": 24, "activ": [78, 84], "ad": [12, 90], "adapt": 53, "addit": 3, "advanc": 59, "alibi": 5, "analysi": 67, "announc": 88, "api": [3, 7, 11, 17, 26, 36, 64, 65, 70, 83, 88, 91], "arbitrari": 3, "architectur": [16, 24, 59, 91], "argument": 25, "asynchron": 42, "asyncio": 32, "attent": [5, 13, 24, 66, 74, 75, 76, 78, 92], "attentionbackend": 92, "attentionmetadata": 92, "auto": 25, "automat": 38, "autoregress": 24, "avoid": 70, "awq": [13, 19, 85], "b200": [18, 24], "backend": [24, 86, 89, 90, 92], "background": 24, "balanc": 24, "base": 32, "baselin": 72, "batch": [3, 5, 66, 74], "beam": [3, 5], "befor": [68, 70], "begin": 70, "behavior": 68, "bench": [51, 67, 70], "benchmark": [2, 18, 23, 26, 68, 69, 70], "best": [18, 23], "bf16": 85, "bia": 5, "bind": [3, 14, 60], "blackwel": 85, "boost": 68, "boundari": 24, "budget": 22, "buffer": [5, 72, 84], "buffermanag": 1, "build": [13, 17, 18, 25, 27, 28, 32, 60, 68, 70, 75], "c": [3, 6, 60, 84], "cach": [5, 8, 13, 46, 72, 76, 84, 93], "cachecommun": 0, "can": [8, 66], "capac": 76, "case": 74, "cast": 78, "caveat": 68, "chang": [11, 74, 88], "chat": [26, 29, 30, 55, 56], "checkpoint": 13, "choos": 23, "chunk": [5, 18, 74, 76], "class": 3, "classic": 7, "cli": [17, 70], "client": [29, 30, 31, 34, 35, 55, 56, 57], "clock": [18, 68], "close": [19, 22], "code": 60, "collect": 67, "combin": 18, "come": 23, "command": 69, "common": [1, 32, 66], "commun": [24, 71], "compil": [14, 18, 60, 83], "complet": [26, 31, 57], "compon": [6, 89], "conclus": [72, 74, 75], "config": [13, 25], "configur": [3, 6, 9, 24, 28, 32, 72, 75, 90], "connect": 28, "consumpt": 11, "contain": [18, 27, 60], "content": [18, 24, 73, 90], "context": [3, 5, 18, 74, 75, 76], "contigu": 5, "control": [3, 47], "conv": 78, "convers": [12, 17], "coordin": 67, "core": 90, "cpp": 9, "creat": [28, 60], "cross": 5, "cuda": 24, "cudaev": 1, "cudastream": 1, "curl": [29, 30, 31], "custom": [15, 32, 44, 93, 94], "cutlass": 24, "cyclic": 5, "dataset": [18, 68, 69, 70], "datatransceiverst": 0, "debug": [2, 67, 87], "decid": 71, "decod": [3, 10, 25, 39, 40, 48, 49, 84, 91], "decoderst": 1, "decodinginput": 1, "decodingoutput": 1, "decor": 7, "deepseek": [18, 24, 33], "default": [18, 24, 68, 70], "definit": [14, 83, 90], "dens": 24, "depend": 24, "deploi": 83, "dequant": 85, "descript": 67, "detail": [9, 85], "develop": [28, 89], "diagram": 24, "differ": 3, "disabl": 32, "disaggreg": [2, 26], "disaggregated_mpi_work": 26, "disaggserverutil": 0, "distribut": [45, 50], "do": 66, "docker": [27, 28, 60], "dockerhub": [27, 28], "document": [59, 88], "dora": 9, "download": 18, "dq": 85, "draft": 10, "e2": 87, "eagl": [10, 39], "eaglebuff": 1, "eaglemodul": 1, "embed": [5, 78], "enabl": [4, 8, 18, 27, 67, 72, 75], "endpoint": 26, "engin": [13, 14, 64, 68, 70, 83, 91], "enhanc": 88, "environ": 2, "error": 87, "etp": 24, "evalu": 13, "event": 46, "everyth": 24, "exampl": [2, 3, 9, 13, 14, 15, 36, 37, 58, 67, 68], "except": 84, "execut": 87, "executor": [0, 3, 9], "expect": [8, 18], "experiment": 2, "expert": [4, 24], "explicitdrafttokensbuff": 1, "explor": 18, "face": 64, "factor": [5, 13], "falcon": 19, "faq": [2, 84], "faster": 19, "featur": [18, 63, 67, 88], "file": 60, "first": 20, "fix": 88, "flag": [75, 85], "flayerinfo": 7, "flight": [3, 5, 66], "flow": 68, "fmha": 5, "format": [9, 18], "fp16": 85, "fp32": 85, "fp4": 69, "fp8": [5, 13, 20, 66, 69, 72, 85], "fraction": 76, "free": 76, "from": 60, "full": 60, "fulli": 15, "function": [7, 15, 77], "fuse_a_gemm": 24, "fusion": [14, 24, 72, 75], "futur": [24, 32], "garbag": 67, "gate": 72, "gc": 67, "gemm": [24, 72, 75], "genai": [34, 35], "gener": [2, 5, 32, 39, 40, 41, 42, 43, 44, 45, 47, 48, 49, 53, 54], "get": [18, 46, 59], "gil": 67, "gpt": [6, 9], "gptdecod": 1, "gptdecoderbatch": 1, "gptjsonconfig": 1, "gptq": 85, "gpu": [11, 14, 18, 19, 24, 66, 68, 76, 84], "grace": 61, "graph": [7, 24], "group": [5, 24], "guid": [3, 40, 73, 83, 89, 90], "h": [0, 1], "h100": [20, 21], "h200": [18, 19, 21, 22], "ha": 20, "hardwar": 86, "hbm": 21, "head": 5, "header": 60, "high": 7, "hopper": [61, 85], "host": 8, "how": [4, 8, 18, 24, 68, 71, 74], "hub": 64, "hug": 64, "i": [20, 71, 84], "ibuff": 1, "id": 9, "igptdecoderbatch": 1, "imag": [27, 28, 60], "implement": [12, 24, 92], "import": 5, "improv": 10, "increas": 22, "indic": 59, "infer": [3, 23, 26, 66, 83, 84], "inform": [7, 67, 83], "infrastructur": 88, "input": 5, "instal": [18, 59, 61, 62, 87], "int4": [19, 85], "int8": [5, 85], "interfac": 93, "intern": 6, "introduct": [36, 64, 90, 93, 94], "ipcnvlsmemori": 1, "ipcutil": 1, "isl": 18, "issu": [18, 84, 88, 89], "itensor": 1, "iter": 67, "kei": [15, 24, 28, 63, 71, 88, 89], "kernel": [22, 24], "knowledg": 73, "known": [60, 84, 88, 89], "kv": [5, 8, 13, 46, 72, 76, 84, 93], "kvcachemanag": 91, "latenc": [18, 22, 24, 68, 70, 72], "latest": [21, 66], "launch": [24, 67], "layer": [24, 78], "layernorm": 13, "layout": 15, "level": [7, 24, 91], "limit": [10, 60, 68, 88], "linear": 78, "link": 60, "linux": [60, 62], "llama": [14, 19, 22, 68, 69, 72, 75], "llama2": 21, "llm": [4, 10, 13, 15, 16, 17, 18, 20, 21, 23, 27, 28, 32, 36, 37, 38, 45, 50, 51, 52, 59, 60, 64, 66, 68, 70, 74, 83, 84, 86, 88], "load": [15, 90], "loader": 15, "local": 64, "logit": [3, 25, 47], "lookahead": [10, 48], "lookaheadbuff": 1, "lookaheadmodul": 1, "lookup": 10, "lora": [9, 25, 53], "loracach": [1, 9], "loracachepagemanagerconfig": 1, "loramodul": 1, "low": [68, 72], "make": 13, "manag": [7, 68, 93], "map": [9, 68], "mark": 3, "marker": 67, "match": 14, "matrix": [85, 86], "max": [18, 68, 74, 76], "maximum": 76, "measur": 69, "medusa": [10, 49, 68], "medusamodul": 1, "memori": [8, 11, 18, 21, 76, 84], "memorycount": 1, "method": [7, 23], "metric": 26, "mgmn": [50, 51, 52], "min": 18, "mix": 24, "mixtur": 4, "mlp": [13, 72, 78], "mlperf": 20, "modal": [68, 86], "mode": 68, "model": [6, 10, 12, 14, 15, 16, 18, 24, 64, 68, 69, 71, 72, 75, 79, 83, 86, 87, 88, 90, 91], "modelconfig": 1, "modul": [7, 9], "moe": 4, "moe_backend": 24, "more": [18, 22, 67], "mtp": 24, "multi": [5, 14, 24, 26, 66, 68, 86], "multimod": [26, 30, 35], "multipl": [53, 75], "name": [15, 25], "nativ": [15, 66], "nearli": 21, "network": 68, "new": [12, 22, 90, 92], "next": [23, 83], "node": [14, 26, 66], "non": 68, "norm": [72, 75], "normal": 78, "note": [3, 5, 88], "nsight": 67, "num": 74, "numer": 85, "nvfp4": 85, "nvidia": [24, 67], "nvtx": 67, "o": 84, "obtain": 3, "offload": 8, "onli": [24, 60, 67, 85], "onlin": 58, "openai": [55, 56, 57], "optim": [5, 24, 75], "option": [18, 60, 72, 75, 76], "osl": 18, "other": 68, "out": [18, 90], "output": [3, 68], "over": 19, "overview": [6, 13, 15, 17, 66, 69], "ovewiew": 91, "own": 94, "p": 8, "pack": 5, "pad": 5, "page": [5, 66, 74, 75, 76], "parallel": [4, 9, 24, 25, 38, 68, 71, 75], "paramet": 6, "parser": 33, "part": 12, "pattern": [7, 14], "perf": [34, 35], "perform": [8, 10, 18, 20, 23, 24, 59, 67, 70, 72, 73, 75], "persist": 68, "phase": 5, "pipelin": [71, 75], "pitfal": 70, "plugin": [14, 25, 72, 75, 80], "pod": 28, "polici": 76, "pool": [78, 84], "posit": 5, "post": 3, "postprocess": 15, "power": 68, "practic": 23, "precis": [24, 85], "prepar": [13, 18, 28, 64, 68, 69, 70], "prerequisit": [18, 60, 73, 83, 90], "prevent": 8, "processor": [3, 47], "profil": [24, 67, 75], "programmat": 24, "prompt": 10, "prompttuningparam": 1, "provid": 22, "push": 24, "pyexecutor": 91, "python": [3, 60, 84], "pytorch": [67, 68, 86, 89, 90], "q": 85, "qkv": 5, "quantiz": [13, 17, 23, 32, 54, 68, 72, 81, 85, 89], "quantmod": 85, "queri": 5, "quick": [83, 89], "quickstart": 68, "r1": [18, 24, 33], "rab": 5, "rank": 13, "rawengin": 1, "re": 24, "reason": 33, "recommend": [72, 75, 84], "record_signatur": 7, "redraft": 10, "reduc": [11, 72, 75], "refer": [12, 59, 65], "regist": 12, "registr": 90, "rel": 5, "relat": [7, 83], "relax": 24, "releas": 88, "reproduc": [18, 24, 69], "request": [1, 3], "requir": 7, "resourcemanag": 91, "respons": 3, "result": [3, 18, 67, 69, 70], "retriev": 7, "reus": 8, "revisit": 74, "rewrit": 7, "right": 23, "roll": 5, "rope": 5, "rotari": 5, "router": 24, "routergemm": 24, "run": [9, 11, 18, 67, 68, 69, 70, 83], "runpod": 28, "runtim": [1, 6, 9, 14, 32, 60, 76, 82, 84], "runtimedefault": 1, "same": 22, "sampl": [6, 10, 32], "samplingconfig": 1, "save": 70, "scale": [5, 13], "scatter": 75, "schedul": [74, 76, 91, 94], "script": [37, 58], "search": 5, "sec": 21, "send": 3, "serial": 0, "serv": [26, 52, 58, 67, 83], "server": [3, 26, 83], "servic": 2, "set": [68, 71], "shard": 71, "shoot": 15, "singl": 19, "situat": 8, "size": [74, 76, 84], "slide": 5, "slurm": 26, "smart": 24, "smoothquant": 85, "softwar": 86, "sota": 23, "sourc": 60, "spars": 24, "specif": 67, "specul": [10, 25], "speculativedecodingmod": 1, "speculativedecodingmodul": 1, "speed": 23, "ssh": [27, 28], "start": [26, 59, 83, 89], "step": [12, 18, 60, 83, 90], "strategi": [24, 71], "stream": [11, 24, 43], "streamingllm": 5, "structur": 3, "studi": 74, "style": 32, "subcommand": 68, "summari": [68, 72, 75], "support": [14, 15, 18, 60, 64, 66, 68, 85, 86], "swiglu": 72, "syntax": 26, "system": [24, 67], "tabl": [18, 24, 59, 73, 90], "target": 10, "technic": 85, "techniqu": 23, "templat": 28, "tensor": [0, 3, 4, 5, 7, 9, 71, 84], "tensorrt": [4, 10, 13, 14, 15, 16, 17, 18, 20, 21, 23, 27, 28, 59, 60, 64, 66, 68, 70, 74, 83, 84, 86, 88], "test": 87, "text": [39, 40, 41, 42, 43, 44, 47, 48, 49, 53], "think": 71, "throughput": [18, 22, 68, 69, 70], "time": [75, 84], "tip": [64, 70, 87], "tllmlogger": 1, "tok": 20, "token": [20, 21, 32, 74, 76], "tool": 17, "top": 91, "translat": 15, "tree": [10, 90], "triton": [3, 83], "troubl": 15, "troubleshoot": [2, 64, 70, 87], "trt": 23, "trtllm": [24, 25, 26, 51, 52, 67, 70, 83], "tune": [8, 18, 73, 74], "type": 0, "understand": [74, 84], "unit": 87, "up": [19, 22, 23], "updat": 88, "upload": [27, 28], "us": [7, 9, 10, 39, 47, 48, 49, 75, 76, 84], "usag": [2, 84], "user": 72, "v": [4, 21], "valid": 68, "variabl": [2, 69], "verif": 24, "verifi": 12, "via": 70, "visual": 67, "w4a16": 85, "w8a16": 85, "w8a8": 85, "weight": [11, 12, 13, 14, 15, 16, 84, 85, 90], "welcom": 59, "what": [20, 23, 66], "when": [7, 24], "width": 3, "window": [5, 66, 76], "wip": 18, "within": 22, "without": 60, "work": [24, 68], "workflow": [7, 15, 17, 67, 68], "workload": 24, "world": 6, "worldconfig": 1, "write": 12, "xqa": [5, 22], "you": [66, 70], "your": 94}}) \ No newline at end of file +Search.setIndex({"alltitles": {"1. Download TensorRT-LLM": [[20, "download-tensorrt-llm"]], "1. Weights size": [[89, "weights-size"]], "2. Activation size": [[89, "activation-size"]], "2. Download the DeepSeek R1 models": [[20, "download-the-deepseek-r1-models"]], "3. Build and run TensorRT-LLM container": [[20, "build-and-run-tensorrt-llm-container"]], "3. I/O tensors": [[89, "i-o-tensors"]], "3.1 Runtime and decoder buffers except KV cache tensor": [[89, "runtime-and-decoder-buffers-except-kv-cache-tensor"]], "3.2 KV cache tensor": [[89, "kv-cache-tensor"]], "4. Compile and Install TensorRT-LLM": [[20, "compile-and-install-tensorrt-llm"]], "5. Optional: Tune GPU clocks": [[20, "optional-tune-gpu-clocks"]], "6. Dataset preparation": [[20, "dataset-preparation"]], "@record_signature to Decorate Functionals Requiring FLayerInfo": [[7, "record-signature-to-decorate-functionals-requiring-flayerinfo"]], "ALiBi": [[5, "alibi"]], "API": [[3, "api"]], "API Changes": [[13, "api-changes"], [93, "api-changes"], [93, "id9"], [93, "id14"], [93, "id19"], [93, "id24"], [93, "id31"], [93, "id36"], [93, "id42"], [93, "id48"], [93, "id54"]], "API Introduction": [[69, null]], "API Reference": [[70, null]], "AWQ Quantization Scaling Factors": [[15, "awq-quantization-scaling-factors"]], "About": [[30, "about"]], "About Speculative Sampling": [[12, "about-speculative-sampling"]], "About TensorRT-LLM": [[71, "about-tensorrt-llm"]], "Accuracy": [[25, "accuracy"]], "Accuracy studies for Relaxed Acceptance": [[27, "accuracy-studies-for-relaxed-acceptance"]], "Achieving speedup with MTP speculative decoding": [[27, "achieving-speedup-with-mtp-speculative-decoding"]], "Acknowledgment": [[26, "acknowledgment"], [27, "acknowledgment"], [28, "acknowledgment"]], "Activation": [[83, "module-tensorrt_llm.layers.activation"]], "Adding a Model": [[14, null]], "Adding a New Model in PyTorch Backend": [[95, null]], "Advanced": [[64, null]], "Algorithm": [[11, "algorithm"]], "Announcements": [[93, "announcements"], [93, "id52"]], "Architecture": [[64, null]], "Architecture Ovewiew": [[96, null]], "Asyncio-Based Generation": [[36, "asyncio-based-generation"]], "Attention": [[83, "module-tensorrt_llm.layers.attention"], [97, null]], "Attention Backends": [[97, "attention-backends"]], "Attention Kernel": [[26, "attention-kernel"]], "Attention Weights": [[15, "attention-weights"]], "Attention for MTP": [[27, "attention-for-mtp"]], "Auto parallel arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-auto-parallel-arguments"]], "Automatic Parallelism with LLM": [[42, null]], "Autoregressive MTP Layers": [[26, "autoregressive-mtp-layers"]], "B200 max-throughput": [[20, "b200-max-throughput"]], "B200 min-latency": [[20, "b200-min-latency"]], "Background": [[26, "background"], [27, "background"]], "Basic Implementation": [[27, "basic-implementation"]], "Beam-Search": [[5, "beam-search"]], "Before Benchmarking": [[73, "before-benchmarking"]], "Before You Begin: TensorRT-LLM LLM-API": [[75, "before-you-begin-tensorrt-llm-llm-api"]], "Benchmark": [[20, "benchmark"], [25, "benchmark"], [30, "benchmark"]], "Benchmarking Default Performance": [[75, null]], "Benchmarking a non-Medusa Low Latency Engine": [[73, "benchmarking-a-non-medusa-low-latency-engine"]], "Benchmarking with trtllm-bench": [[75, "benchmarking-with-trtllm-bench"]], "Benchmarks": [[2, "benchmarks"]], "Best practices to choose the right quantization methods": [[25, "best-practices-to-choose-the-right-quantization-methods"]], "Block": [[8, "block"]], "Boost settings": [[73, "boost-settings"]], "Build APIs": [[19, "build-apis"]], "Build Checkpoint into TensorRT Engine": [[15, "build-checkpoint-into-tensorrt-engine"]], "Build Configuration": [[36, "build-configuration"]], "Build TensorRT-LLM": [[65, "build-tensorrt-llm"]], "Build the TensorRT-LLM Docker Image": [[31, null]], "Build the TensorRT-LLM Docker Image and Upload to DockerHub": [[31, "build-the-tensorrt-llm-docker-image-and-upload-to-dockerhub"], [32, "build-the-tensorrt-llm-docker-image-and-upload-to-dockerhub"]], "Building a Benchmark Engine": [[73, "building-a-benchmark-engine"]], "Building a Medusa Low-Latency Engine": [[73, "building-a-medusa-low-latency-engine"]], "Building a TensorRT-LLM Docker Image": [[65, "building-a-tensorrt-llm-docker-image"]], "Building and Saving Engines via CLI": [[75, "building-and-saving-engines-via-cli"]], "Building and Saving the Engine": [[75, "building-and-saving-the-engine"]], "Building from Source Code on Linux": [[65, null]], "Building the Python Bindings for the C++ Runtime": [[65, "building-the-python-bindings-for-the-c-runtime"]], "C++ Executor API Example": [[3, "c-executor-api-example"]], "C++ GPT Runtime": [[6, null]], "C++ runtime": [[89, "c-runtime"], [89, "id1"]], "CLI Tools": [[19, "cli-tools"]], "CUDA Graph & Programmatic Dependent Launch": [[26, "cuda-graph-programmatic-dependent-launch"]], "CUTLASS Backend (default backend)": [[26, "cutlass-backend-default-backend"]], "Capacity Scheduler Policy": [[81, "capacity-scheduler-policy"]], "Cast": [[83, "module-tensorrt_llm.layers.cast"]], "Chat API": [[30, "chat-api"]], "Chunked Context": [[5, "chunked-context"]], "Classical Workflow": [[7, "classical-workflow"]], "Closing": [[21, "closing"], [24, "closing"]], "Collect PyTorch profiler results": [[72, "collect-pytorch-profiler-results"]], "Command Overview": [[74, "command-overview"]], "Common LLM Support": [[71, "common-llm-support"]], "Communication Kernel": [[26, "communication-kernel"]], "Compilation": [[16, "compilation"]], "Compile the Model into a TensorRT Engine": [[88, "compile-the-model-into-a-tensorrt-engine"]], "Completions API": [[30, "completions-api"], [30, "id1"]], "Conclusion": [[77, "conclusion"], [79, "conclusion"], [80, "conclusion"]], "Config": [[15, "config"]], "Configure SSH Key": [[32, "configure-ssh-key"]], "Configure The Executor": [[3, "configure-the-executor"]], "Connect to the Pod": [[32, "connect-to-the-pod"]], "Context Chunking Policy": [[81, "context-chunking-policy"]], "Context Phase": [[5, "context-phase"]], "Context and Generation Phases": [[5, "context-and-generation-phases"]], "Contiguous KV Cache": [[5, "contiguous-kv-cache"]], "Control generated text using logits processor": [[52, null]], "Controlling output with Logits Post-Processor": [[3, "controlling-output-with-logits-post-processor"]], "Conv": [[83, "module-tensorrt_llm.layers.conv"]], "Conversion APIs": [[19, "conversion-apis"]], "Coordinating with NVIDIA Nsight Systems Launch": [[72, "coordinating-with-nvidia-nsight-systems-launch"]], "Coordinating with PyTorch profiler (PyTorch workflow only)": [[72, "coordinating-with-pytorch-profiler-pytorch-workflow-only"]], "Core Models": [[95, "core-models"]], "Create a Pod Template": [[32, "create-a-pod-template"]], "Create a Runpod account": [[32, "create-a-runpod-account"]], "Create the Container": [[65, "create-the-container"]], "Cross Attention": [[5, "cross-attention"]], "Curl Chat Client": [[33, null]], "Curl Chat Client For Multimodal": [[34, null]], "Curl Completion Client": [[35, null]], "Customize KV Cache Manager": [[98, "customize-kv-cache-manager"]], "Customize Your Own Scheduler": [[99, "customize-your-own-scheduler"]], "Data Parallel for Attention module (ADP)": [[28, "data-parallel-for-attention-module-adp"]], "Debug Execution Errors": [[92, "debug-execution-errors"]], "Debug on E2E Models": [[92, "debug-on-e2e-models"]], "Debug on Unit Tests": [[92, "debug-on-unit-tests"]], "Debugging FAQs": [[2, "debugging-faqs"]], "Deciding Model Sharding Strategy": [[76, null]], "Decoder": [[96, "decoder"]], "DeepSeek R1 MTP Implementation and Optimization": [[27, null]], "Deepseek R1 Reasoning Parser": [[37, null]], "Default Build Behavior": [[73, "default-build-behavior"]], "Dense GEMM optimization": [[26, "dense-gemm-optimization"]], "Deploy with Triton Inference Server": [[88, "deploy-with-triton-inference-server"]], "Deploy with trtllm-serve": [[88, "deploy-with-trtllm-serve"]], "Develop TensorRT-LLM on Runpod": [[32, null]], "Developer Guide": [[94, "developer-guide"]], "Disable Tokenizer": [[36, "disable-tokenizer"]], "Disaggregated-Service (experimental)": [[2, null]], "Distributed LLM Generation": [[50, null]], "DoRA": [[10, "dora"]], "Documentation": [[93, "documentation"], [93, "id28"]], "Draft-Target-Model": [[12, "draft-target-model"]], "EAGLE": [[12, "eagle"]], "Eagle3 support": [[27, "eagle3-support"]], "Embedding": [[83, "module-tensorrt_llm.layers.embedding"]], "Enable GIL information in NVTX markers": [[72, "enable-gil-information-in-nvtx-markers"]], "Enable garbage collection (GC) NVTX markers": [[72, "enable-garbage-collection-gc-nvtx-markers"]], "Enable kv cache reuse for p-tuning": [[9, "enable-kv-cache-reuse-for-p-tuning"]], "Enable more NVTX markers for debugging": [[72, "enable-more-nvtx-markers-for-debugging"]], "Enable ssh access to the container": [[31, "enable-ssh-access-to-the-container"]], "Enabling GEMM + SwiGLU Fusion": [[77, "enabling-gemm-swiglu-fusion"]], "Enabling GEMM Plugin": [[80, "enabling-gemm-plugin"]], "Enabling Low Latency GEMM plugin": [[77, "enabling-low-latency-gemm-plugin"]], "Enabling Paged Context Attention": [[80, "enabling-paged-context-attention"]], "Enabling Quantization": [[77, "enabling-quantization"]], "Enabling Quantized KV Cache": [[77, "enabling-quantized-kv-cache"]], "Enabling Reduce Norm Fusion Plugin": [[80, "enabling-reduce-norm-fusion-plugin"]], "Enabling Reduce Norm Fusion with User Buffers": [[77, "enabling-reduce-norm-fusion-with-user-buffers"]], "Enabling building with multiple profiles": [[80, "enabling-building-with-multiple-profiles"]], "Environment Variables": [[2, "environment-variables"], [11, "environment-variables"]], "Evaluation": [[27, "evaluation"]], "Events in KVCacheEventManager": [[8, "events-in-kvcacheeventmanager"]], "Everything in One Diagram": [[26, "everything-in-one-diagram"]], "Example": [[2, "example"], [15, "example"]], "Example LoRA tensors": [[10, "example-lora-tensors"]], "Example of Build Subcommand Output:": [[73, "example-of-build-subcommand-output"]], "Examples": [[16, "examples"], [17, "examples"], [72, "examples"]], "Executor": [[0, null]], "Executor API": [[3, null]], "Expected Result Format": [[20, "expected-result-format"], [20, "id1"], [20, "id2"]], "Expected Results": [[20, "expected-results"]], "Expert Parallelism in TensorRT-LLM": [[4, null]], "Expert parallel for MoE (EP)": [[28, "expert-parallel-for-moe-ep"]], "Exploring more ISL/OSL combinations": [[20, "exploring-more-isl-osl-combinations"]], "FAQ": [[89, "faq"]], "FLayerInfo for Retrieving High-Level Information for a Functional": [[7, "flayerinfo-for-retrieving-high-level-information-for-a-functional"]], "FP32, FP16 and BF16": [[90, "fp32-fp16-and-bf16"]], "FP4 Models:": [[74, "fp4-models"]], "FP8 (Hopper)": [[90, "fp8-hopper"]], "FP8 Context FMHA": [[5, "fp8-context-fmha"]], "FP8 Models:": [[74, "fp8-models"]], "FP8 Quantization": [[77, null]], "FP8 Quantization Scaling Factors": [[15, "fp8-quantization-scaling-factors"]], "FP8 Support": [[71, "fp8-support"]], "FP8 \u201cBaseline\u201d Performance": [[77, "fp8-baseline-performance"]], "Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100": [[21, null]], "Falcon-180B on a single H200 with INT4 AWQ": [[21, "falcon-180b-on-a-single-h200-with-int4-awq"]], "Feature Descriptions": [[72, "feature-descriptions"]], "Fix known issues": [[27, "fix-known-issues"]], "Fixed Issues": [[93, "fixed-issues"], [93, "id11"], [93, "id15"], [93, "id21"], [93, "id26"], [93, "id33"], [93, "id38"], [93, "id44"], [93, "id50"], [93, "id56"], [93, "id61"]], "Fully customized": [[17, "fully-customized"]], "Functionals": [[82, null]], "Fuse_A_GEMM": [[26, "fuse-a-gemm"]], "Future Works": [[26, "future-works"], [27, "future-works"], [28, "future-works"]], "Future-Style Generation": [[36, "future-style-generation"]], "GEMM + SwiGLU Fusion in Gated-MLP": [[77, "gemm-swiglu-fusion-in-gated-mlp"]], "GEMM Plugin": [[80, "gemm-plugin"]], "GPTQ and AWQ (W4A16)": [[90, "gptq-and-awq-w4a16"]], "GPU Clock Management": [[73, "gpu-clock-management"]], "Genai Perf Client": [[38, null]], "Genai Perf Client For Multimodal": [[39, null]], "General FAQs": [[2, "general-faqs"]], "Generate Text Asynchronously": [[47, null]], "Generate Text Using Eagle Decoding": [[44, null]], "Generate Text Using Eagle2 Decoding": [[43, null]], "Generate Text Using Lookahead Decoding": [[53, null]], "Generate Text Using Medusa Decoding": [[54, null]], "Generate Text in Streaming": [[48, null]], "Generate text": [[46, null]], "Generate text with customization": [[49, null]], "Generate text with guided decoding": [[45, null]], "Generate text with multiple LoRA adapters": [[58, null]], "Generation": [[36, "generation"]], "Generation Phase": [[5, "generation-phase"]], "Generation with Quantization": [[59, null]], "Get KV Cache Events": [[51, null]], "Getting Started": [[64, null]], "Graph Rewriting APIs": [[7, "graph-rewriting-apis"]], "Graph Rewriting Module": [[7, null]], "Grouped GEMM": [[26, "grouped-gemm"]], "H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token": [[22, null]], "H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLM": [[23, null]], "H200 max-throughput": [[20, "h200-max-throughput"]], "H200 min-latency": [[20, "h200-min-latency"]], "H200 vs H100": [[23, "h200-vs-h100"]], "Hardware": [[91, "hardware"]], "Hierarchy: Pool, Block, and Page": [[8, "hierarchy-pool-block-and-page"]], "How the Benchmarker Works": [[73, "how-the-benchmarker-works"]], "How to Enable": [[4, "how-to-enable"]], "How to Think about Model Sharding: Communication is Key": [[76, "how-to-think-about-model-sharding-communication-is-key"]], "How to change Max Batch Size": [[79, "how-to-change-max-batch-size"]], "How to change Max Num Tokens": [[79, "how-to-change-max-num-tokens"]], "How to enable kv cache reuse": [[9, "how-to-enable-kv-cache-reuse"]], "How to get best performance on DeepSeek-R1 in TensorRT-LLM": [[20, null]], "How to reproduce": [[26, "how-to-reproduce"], [28, "how-to-reproduce"]], "How to run DeepSeek models with MTP": [[27, "how-to-run-deepseek-models-with-mtp"]], "How to run the DeepSeek-R1 model with Relaxed Acceptance": [[27, "how-to-run-the-deepseek-r1-model-with-relaxed-acceptance"]], "How to set Tensor Parallelism and Pipeline Parallelism": [[76, "how-to-set-tensor-parallelism-and-pipeline-parallelism"]], "Hugging Face Hub": [[69, "hugging-face-hub"]], "INT4 and INT8 Weight-Only (W4A16 and W8A16)": [[90, "int4-and-int8-weight-only-w4a16-and-w8a16"]], "INT8 SmoothQuant (W8A8)": [[90, "int8-smoothquant-w8a8"]], "INT8/FP8 KV Caches": [[5, "int8-fp8-kv-caches"]], "Implement AttentionBackend": [[97, "implement-attentionbackend"]], "Implement AttentionMetadata": [[97, "implement-attentionmetadata"]], "Implement a New Attention Backend": [[97, "implement-a-new-attention-backend"]], "Implementation Configuration": [[26, "implementation-configuration"]], "Important Note": [[5, "important-note"]], "In-Flight Batching and Paged Attention": [[71, "in-flight-batching-and-paged-attention"]], "In-flight Batching": [[5, "in-flight-batching"]], "In-flight Batching with the Triton Inference Server": [[3, "in-flight-batching-with-the-triton-inference-server"]], "Indices and tables": [[64, "indices-and-tables"]], "Inference Endpoints": [[30, "inference-endpoints"]], "Infrastructure Changes": [[93, "infrastructure-changes"], [93, "id4"], [93, "id7"], [93, "id12"], [93, "id16"], [93, "id22"], [93, "id27"], [93, "id34"], [93, "id39"], [93, "id45"]], "Infrastructure changes": [[93, "id51"]], "Input QKV tensor": [[5, "input-qkv-tensor"]], "Installation": [[64, null]], "Installation Errors": [[92, "installation-errors"]], "Installing on Grace Hopper": [[66, null]], "Installing on Linux": [[67, null]], "Interfaces": [[98, "interfaces"]], "Internal Components": [[6, "internal-components"]], "Introduction": [[28, "introduction"], [95, "introduction"]], "KV Cache": [[5, "kv-cache"]], "KV Cache Management: Pools, Blocks, and Events": [[8, null]], "KV Cache Manager": [[98, null]], "KV Cache Manager Introduction": [[98, "kv-cache-manager-introduction"]], "KV Cache Pool Management": [[8, "kv-cache-pool-management"]], "KV Cache Quantization Scaling Factors": [[15, "kv-cache-quantization-scaling-factors"]], "KV cache reuse": [[9, null]], "KVCacheManager": [[96, "kvcachemanager"]], "Kernel Level optimizations": [[26, "kernel-level-optimizations"]], "Kernel fusion": [[26, "kernel-fusion"]], "Key Components": [[94, "key-components"]], "Key Features": [[68, null]], "Key Features and Enhancements": [[93, "key-features-and-enhancements"], [93, "id2"], [93, "id3"], [93, "id5"], [93, "id8"], [93, "id13"], [93, "id18"], [93, "id23"], [93, "id30"], [93, "id35"], [93, "id41"], [93, "id47"], [93, "id53"], [93, "id57"], [93, "id59"]], "Key Optimizations": [[26, "key-optimizations"]], "Known Issues": [[89, "known-issues"], [93, "known-issues"], [93, "id6"], [93, "id10"], [93, "id17"], [93, "id29"], [93, "id40"], [93, "id46"], [93, "id62"], [94, "known-issues"]], "Known Limitations": [[65, "known-limitations"]], "LLM API": [[88, "llm-api"]], "LLM API Examples": [[40, null]], "LLM Common Customizations": [[36, null]], "LLM Examples": [[41, null]], "LLM Examples Introduction": [[40, null]], "LLM Models": [[91, "llm-models"]], "Latest GPU Support": [[71, "latest-gpu-support"]], "Latest HBM Memory": [[23, "latest-hbm-memory"]], "LayerNorm Weights": [[15, "layernorm-weights"]], "Layers": [[83, null]], "Limitations": [[12, "limitations"], [93, "limitations"]], "Limitations and Caveats": [[73, "limitations-and-caveats"]], "Linear": [[83, "module-tensorrt_llm.layers.linear"]], "Linking with the TensorRT-LLM C++ Runtime": [[65, "linking-with-the-tensorrt-llm-c-runtime"]], "Llama 3.1 405B": [[16, "llama-3-1-405b"]], "Llama 3.1 405B FP4": [[74, "llama-3-1-405b-fp4"]], "Llama 3.1 405B FP8": [[74, "llama-3-1-405b-fp8"]], "Llama 3.1 70B": [[16, "llama-3-1-70b"]], "Llama 3.1 70B FP8": [[74, "llama-3-1-70b-fp8"]], "Llama 3.1 8B FP8": [[74, "llama-3-1-8b-fp8"]], "Llama 3.3 70B FP4": [[74, "llama-3-3-70b-fp4"]], "Llama-70B on H200 up to 2.4x increased throughput with XQA within same latency budget": [[24, "llama-70b-on-h200-up-to-2-4x-increased-throughput-with-xqa-within-same-latency-budget"]], "Llama-70B on H200 up to 6.7x A100": [[21, "llama-70b-on-h200-up-to-6-7x-a100"]], "Llm Mgmn Llm Distributed": [[55, null]], "Llm Mgmn Trtllm Bench": [[56, null]], "Llm Mgmn Trtllm Serve": [[57, null]], "LoRA Module id mapping": [[10, "lora-module-id-mapping"]], "LoRA arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-lora-arguments"]], "LoRA tensor format details": [[10, "lora-tensor-format-details"]], "LoRA with tensor parallel": [[10, "lora-with-tensor-parallel"]], "Loading function": [[17, "loading-function"]], "Local Hugging Face Models": [[69, "local-hugging-face-models"]], "Local TensorRT-LLM Engine": [[69, "local-tensorrt-llm-engine"]], "Logits arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-logits-arguments"]], "Lookahead Decoding": [[12, "lookahead-decoding"]], "LoraCache configuration": [[10, "loracache-configuration"]], "Low Latency Benchmark": [[73, "low-latency-benchmark"]], "Low Latency GEMM Plugin": [[77, "low-latency-gemm-plugin"]], "Low Latency TensorRT-LLM Engine for Llama-3 70B": [[73, "low-latency-tensorrt-llm-engine-for-llama-3-70b"]], "Low-Precision-AllReduce": [[11, null]], "MLA Layers Optimizations": [[28, "mla-layers-optimizations"]], "MLP": [[83, "module-tensorrt_llm.layers.mlp"]], "MLP Weights": [[15, "mlp-weights"]], "MLPerf on H100 with FP8": [[22, "mlperf-on-h100-with-fp8"]], "MTP": [[26, "mtp"]], "MTP Eagle": [[27, "mtp-eagle"]], "MTP Modules": [[27, "mtp-modules"]], "MTP Vanilla": [[27, "mtp-vanilla"]], "MTP for inference": [[27, "mtp-for-inference"]], "MTP implementation in TensorRT-LLM": [[27, "mtp-implementation-in-tensorrt-llm"]], "MTP optimization - Relaxed Acceptance": [[27, "mtp-optimization-relaxed-acceptance"]], "Make Evaluation": [[15, "make-evaluation"]], "Mark Tensors As Output": [[3, "mark-tensors-as-output"]], "Max Throughput Benchmark": [[73, "max-throughput-benchmark"]], "Max Tokens in Paged KV Cache and KV Cache Free GPU Memory Fraction": [[81, "max-tokens-in-paged-kv-cache-and-kv-cache-free-gpu-memory-fraction"]], "Maximum Attention Window Size": [[81, "maximum-attention-window-size"]], "Medusa": [[12, "medusa"]], "Medusa Tree": [[12, "medusa-tree"]], "Memory Usage of TensorRT-LLM": [[89, null]], "Memory pool": [[89, "memory-pool"]], "Metrics Endpoint": [[30, "metrics-endpoint"]], "Mixed ETP": [[26, "mixed-etp"]], "Mixture of Experts (MoE)": [[4, "mixture-of-experts-moe"]], "MoE Layers Optimizations": [[28, "moe-layers-optimizations"]], "Model Architecture": [[26, "model-architecture"]], "Model Configuration": [[6, "model-configuration"], [95, "model-configuration"]], "Model Definition": [[16, null], [95, "model-definition"]], "Model Definition API": [[88, "model-definition-api"]], "Model Engine": [[16, "model-engine"], [96, "model-engine"]], "Model Preparation": [[69, "model-preparation"]], "Model Registration": [[95, "model-registration"]], "Model Updates": [[93, "model-updates"], [93, "id20"], [93, "id25"], [93, "id32"], [93, "id37"], [93, "id43"], [93, "id49"], [93, "id55"], [93, "id58"], [93, "id60"]], "Model Weights": [[18, "model-weights"]], "Models": [[84, null]], "Models (PyTorch Backend)": [[91, "models-pytorch-backend"]], "Models (TensorRT Backend)": [[91, "models-tensorrt-backend"]], "Models with customized key names": [[17, "models-with-customized-key-names"]], "Models with customized weight layout": [[17, "models-with-customized-weight-layout"]], "Multi-GPU Multi-Node Inference": [[71, "multi-gpu-multi-node-inference"]], "Multi-GPU and Multi-Node Support": [[16, "multi-gpu-and-multi-node-support"]], "Multi-Head, Multi-Query, and Group-Query Attention": [[5, null]], "Multi-Modal Models 3": [[91, "multi-modal-models"]], "Multi-node Serving with Slurm": [[30, "multi-node-serving-with-slurm"]], "Multi-streams": [[26, "multi-streams"]], "Multimodal Serving": [[30, "multimodal-serving"]], "Multiple Profiles": [[80, "multiple-profiles"]], "NVFP4 (Blackwell)": [[90, "nvfp4-blackwell"]], "Named Arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-named-arguments"]], "Native Windows Support": [[71, "native-windows-support"]], "Natively supported models": [[17, "natively-supported-models"]], "New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget": [[24, null]], "Next Steps": [[88, "next-steps"]], "Normalization": [[83, "module-tensorrt_llm.layers.normalization"]], "Note on context outputs": [[3, "note-on-context-outputs"]], "Numerical Precision": [[90, null]], "Obtaining Arbitrary Output Tensors": [[3, "obtaining-arbitrary-output-tensors"]], "Offloading to host memory": [[9, "offloading-to-host-memory"]], "Online Serving Examples": [[63, null]], "Only collect specific iterations": [[72, "only-collect-specific-iterations"]], "OpenAI Chat Client": [[60, null], [61, null]], "OpenAI Completion Client": [[62, null]], "Optimizing DeepSeek R1 Throughput on NVIDIA Blackwell GPUs: A Deep Dive for Developers": [[28, null]], "Option 1: Build TensorRT-LLM in One Step": [[65, "option-1-build-tensorrt-llm-in-one-step"]], "Option 1: Full Build with C++ Compilation": [[65, "option-1-full-build-with-c-compilation"]], "Option 2: Build TensorRT-LLM Step-by-Step": [[65, "option-2-build-tensorrt-llm-step-by-step"]], "Option 2: Python-Only Build without C++ Compilation": [[65, "option-2-python-only-build-without-c-compilation"]], "Other Build Modes": [[73, "other-build-modes"]], "Out of memory issues": [[20, "out-of-memory-issues"]], "Out-of-Tree Models": [[95, "out-of-tree-models"]], "Overview": [[6, "overview"], [15, "overview"], [17, "overview"], [19, "overview"], [71, null], [74, null]], "Padded and Packed Tensors": [[5, "padded-and-packed-tensors"]], "Page": [[8, "page"]], "Paged Context Attention": [[80, "paged-context-attention"]], "Paged KV Cache": [[5, "paged-kv-cache"]], "Parallel strategy": [[28, "parallel-strategy"]], "Parallelism Mapping Support": [[73, "parallelism-mapping-support"]], "Parallelism Strategy": [[26, "parallelism-strategy"]], "Pattern and Pattern Manager": [[7, "pattern-and-pattern-manager"]], "Pattern-Matching and Fusion": [[16, "pattern-matching-and-fusion"]], "Performance": [[25, "performance"], [64, null], [80, "performance"]], "Performance Analysis": [[72, null]], "Performance Improvements": [[12, "performance-improvements"]], "Performance Tuning Guide": [[78, null]], "Performance and Accuracy Considerations": [[11, "performance-and-accuracy-considerations"]], "Performance expectations": [[9, "performance-expectations"]], "Performance with GEMM + SwiGLU Fusion": [[77, "performance-with-gemm-swiglu-fusion"]], "Performance with GEMM Plugin": [[80, "performance-with-gemm-plugin"]], "Performance with Low Latency GEMM plugin": [[77, "performance-with-low-latency-gemm-plugin"]], "Performance with Quantized KV Cache": [[77, "performance-with-quantized-kv-cache"]], "Performance with Reduce Norm Fusion": [[80, "performance-with-reduce-norm-fusion"]], "Performance with Reduce Norm Fusion + User Buffers:": [[77, "performance-with-reduce-norm-fusion-user-buffers"]], "Performance with multiple profiles": [[80, "performance-with-multiple-profiles"]], "Persistence mode": [[73, "persistence-mode"]], "Pipeline Parallel Reduce Scatter Optimization": [[80, "pipeline-parallel-reduce-scatter-optimization"]], "Plugin": [[85, null]], "Plugin config arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-plugin-config-arguments"]], "Plugins": [[16, "plugins"]], "Pool": [[8, "pool"]], "Pooling": [[83, "module-tensorrt_llm.layers.pooling"]], "Postprocessing functions": [[17, "postprocessing-functions"]], "Precision Strategy": [[26, "precision-strategy"]], "Precision strategy": [[28, "precision-strategy"]], "Prepare": [[32, "prepare"]], "Prepare Dataset": [[75, "prepare-dataset"]], "Prepare the TensorRT-LLM Checkpoint": [[15, "prepare-the-tensorrt-llm-checkpoint"]], "Preparing a Dataset": [[73, "preparing-a-dataset"], [74, "preparing-a-dataset"]], "Prerequisite Knowledge": [[78, "prerequisite-knowledge"]], "Prerequisites": [[65, "prerequisites"], [88, "prerequisites"], [95, "prerequisites"]], "Prerequisites: Install TensorRT-LLM and download models": [[20, "prerequisites-install-tensorrt-llm-and-download-models"]], "Profiling specific iterations on a trtllm-bench/trtllm-serve run": [[72, "profiling-specific-iterations-on-a-trtllm-bench-trtllm-serve-run"]], "Prompt-Lookup-Decoding": [[12, "prompt-lookup-decoding"]], "Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs": [[26, null]], "PyExecutor": [[96, "pyexecutor"]], "PyTorch Backend": [[94, null]], "Python Bindings for the Executor API": [[3, "python-bindings-for-the-executor-api"]], "Python runtime (Not recommended to be used)": [[89, "python-runtime-not-recommended-to-be-used"]], "Quantization": [[36, "quantization"], [86, null], [94, "quantization"]], "Quantization APIs": [[19, "quantization-apis"]], "Quantization and Dequantization (Q/DQ)": [[90, "quantization-and-dequantization-q-dq"]], "Quantization in TensorRT-LLM": [[25, "quantization-in-tensorrt-llm"]], "Quantization in the PyTorch Flow": [[73, "quantization-in-the-pytorch-flow"]], "Quantized KV-Cache": [[77, "quantized-kv-cache"]], "Quick Start": [[94, "quick-start"]], "Quick Start Guide": [[88, null]], "Quickstart": [[73, "quickstart"]], "Rank Weights": [[15, "rank-weights"]], "Re-balanced the sparse experts": [[26, "re-balanced-the-sparse-experts"]], "ReDrafter": [[12, "redrafter"]], "Reduce Norm Fusion Plugin for Llama models:": [[80, "reduce-norm-fusion-plugin-for-llama-models"]], "Reduce Norm Fusion with User Buffers for Llama Models": [[77, "reduce-norm-fusion-with-user-buffers-for-llama-models"]], "Reference": [[14, "reference"], [64, null]], "Related Information": [[88, "related-information"]], "Relative Attention Bias (RAB)": [[5, "relative-attention-bias-rab"]], "Relax Acceptance Verification": [[26, "relax-acceptance-verification"]], "Relaxed Acceptance": [[27, "relaxed-acceptance"]], "Release Notes": [[93, null]], "Reproducing Benchmarked Results": [[74, "reproducing-benchmarked-results"]], "Reproducing steps": [[20, "reproducing-steps"]], "Request Additional Output": [[3, "request-additional-output"]], "ResourceManager": [[96, "resourcemanager"]], "Results": [[75, "results"]], "Revisiting Paged Context Attention and Context Chunking": [[79, "revisiting-paged-context-attention-and-context-chunking"]], "Rotary Positional Embedding (RoPE)": [[5, "rotary-positional-embedding-rope"]], "RouterGEMM": [[26, "routergemm"]], "Run gpt-2b + LoRA using Executor / cpp runtime": [[10, null]], "Run the Model": [[88, "run-the-model"]], "Running Throughput and Latency Benchmarks": [[75, "running-throughput-and-latency-benchmarks"]], "Running With Weight Streaming to Reduce GPU Memory Consumption": [[13, null]], "Running multi-modal models in the PyTorch Workflow": [[73, "running-multi-modal-models-in-the-pytorch-workflow"]], "Running the Benchmark": [[74, "running-the-benchmark"]], "Running with the PyTorch Workflow": [[73, "running-with-the-pytorch-workflow"]], "Runtime": [[1, null], [16, "runtime"], [87, null]], "Runtime Customization": [[36, "runtime-customization"]], "Runtime Optimizations": [[28, "runtime-optimizations"]], "Sampling": [[36, "sampling"]], "Sampling Parameters": [[6, "sampling-parameters"]], "Scaling factor(s)": [[5, "scaling-factor-s"]], "Scheduler": [[96, "scheduler"], [99, null]], "Scheduler Introduction": [[99, "scheduler-introduction"]], "Scripts": [[41, null], [63, null]], "Sending Requests with Different Beam Widths": [[3, "sending-requests-with-different-beam-widths"]], "Set power limits": [[73, "set-power-limits"]], "Situations that can prevent kv cache reuse": [[9, "situations-that-can-prevent-kv-cache-reuse"]], "Sliding Window Attention, Cyclic (Rolling Buffer) KV Cache": [[5, "sliding-window-attention-cyclic-rolling-buffer-kv-cache"]], "Smart Router": [[26, "smart-router"]], "Software": [[91, "software"]], "Sparse Experts as GEMMs (only works when moe_backend=CUTLASS)": [[26, "sparse-experts-as-gemms-only-works-when-moe-backend-cutlass"]], "Speculative Sampling": [[12, null]], "Speculative decoding arguments": [[29, "tensorrt_llm.commands.build-parse_arguments-speculative-decoding-arguments"]], "Speed up inference with SOTA quantization techniques in TRT-LLM": [[25, null]], "Starting a Server": [[30, "starting-a-server"]], "Step 1. Write Modeling Part": [[14, "step-1-write-modeling-part"]], "Step 2. Implement Weight Conversion": [[14, "step-2-implement-weight-conversion"]], "Step 3. Register New Model": [[14, "step-3-register-new-model"]], "Step 4. Verify New Model": [[14, "step-4-verify-new-model"]], "Step-by-Step Guide": [[95, "step-by-step-guide"]], "StreamingLLM": [[5, "streamingllm"]], "Structured output with guided decoding": [[3, "structured-output-with-guided-decoding"]], "Summary": [[73, "summary"]], "Summary of Configuration Option Recommendations:": [[77, "summary-of-configuration-option-recommendations"], [80, "summary-of-configuration-option-recommendations"]], "Support Matrix": [[91, null]], "Support matrix": [[90, "support-matrix"]], "Supported C++ Header Files": [[65, "supported-c-header-files"]], "Supported Models": [[69, "supported-models"]], "Supported Quantization Modes": [[73, "supported-quantization-modes"]], "Syntax": [[30, "syntax"]], "System Level optimizations": [[26, "system-level-optimizations"]], "TRTLLM Backend": [[26, "trtllm-backend"]], "Table of Contents": [[20, "table-of-contents"], [26, "table-of-contents"], [27, "table-of-contents"], [28, "table-of-contents"], [78, "table-of-contents"], [95, "table-of-contents"]], "Technical Detail: The QuantMode Flags": [[90, "technical-detail-the-quantmode-flags"]], "Tensor Parallel vs Expert Parallel": [[4, "tensor-parallel-vs-expert-parallel"]], "Tensor-Related Methods": [[7, "tensor-related-methods"]], "TensorRT Compiler": [[16, "tensorrt-compiler"]], "TensorRT-LLM Architecture": [[18, null]], "TensorRT-LLM Benchmarking": [[73, null]], "TensorRT-LLM Build Workflow": [[19, null]], "TensorRT-LLM Checkpoint": [[15, null]], "TensorRT-LLM Model Weights Loader": [[17, null]], "TensorRT-LLM Release 0.10.0": [[93, "tensorrt-llm-release-0-10-0"]], "TensorRT-LLM Release 0.11.0": [[93, "tensorrt-llm-release-0-11-0"]], "TensorRT-LLM Release 0.12.0": [[93, "tensorrt-llm-release-0-12-0"]], "TensorRT-LLM Release 0.13.0": [[93, "tensorrt-llm-release-0-13-0"]], "TensorRT-LLM Release 0.14.0": [[93, "tensorrt-llm-release-0-14-0"]], "TensorRT-LLM Release 0.15.0": [[93, "tensorrt-llm-release-0-15-0"]], "TensorRT-LLM Release 0.16.0": [[93, "tensorrt-llm-release-0-16-0"]], "TensorRT-LLM Release 0.17.0": [[93, "tensorrt-llm-release-0-17-0"]], "TensorRT-LLM Release 0.18.0": [[93, "tensorrt-llm-release-0-18-0"]], "TensorRT-LLM Release 0.18.1": [[93, "tensorrt-llm-release-0-18-1"]], "TensorRT-LLM Release 0.18.2": [[93, "tensorrt-llm-release-0-18-2"]], "TensorRT-LLM Release 0.19.0": [[93, "tensorrt-llm-release-0-19-0"]], "TensorRT-LLM Release 0.7.1": [[93, "tensorrt-llm-release-0-7-1"]], "TensorRT-LLM Release 0.8.0": [[93, "tensorrt-llm-release-0-8-0"]], "TensorRT-LLM Release 0.9.0": [[93, "tensorrt-llm-release-0-9-0"]], "The Executor Class": [[3, "the-executor-class"]], "The Request Class": [[3, "the-request-class"]], "The Response Class": [[3, "the-response-class"]], "The Result Class": [[3, "the-result-class"]], "Throughput Benchmarking": [[73, "throughput-benchmarking"]], "Throughput Measurements": [[74, "throughput-measurements"]], "Tips": [[92, "tips"]], "Tips and Troubleshooting": [[69, "tips-and-troubleshooting"]], "Tokenizer Customization": [[36, "tokenizer-customization"]], "Top Level API": [[96, "top-level-api"]], "Topology Requirements": [[11, "topology-requirements"]], "Translator": [[17, "translator"]], "Tree-based speculative decoding support": [[27, "tree-based-speculative-decoding-support"]], "Trouble shooting": [[17, "trouble-shooting"]], "Troubleshooting": [[92, null]], "Troubleshooting Tips and Pitfalls To Avoid": [[75, "troubleshooting-tips-and-pitfalls-to-avoid"]], "Troubleshooting and FAQ": [[2, "troubleshooting-and-faq"]], "Tuning Case Study": [[79, "tuning-case-study"], [79, "id2"]], "Tuning Max Batch Size": [[79, "tuning-max-batch-size"]], "Tuning Max Batch Size and Max Num Tokens": [[79, null]], "Tuning Max Num Tokens": [[79, "tuning-max-num-tokens"]], "Types of Events": [[8, "types-of-events"]], "Understand inference time GPU memory usage": [[89, "understand-inference-time-gpu-memory-usage"]], "Understanding the TensorRT-LLM scheduler": [[79, "understanding-the-tensorrt-llm-scheduler"]], "Upload the Docker Image to DockerHub": [[31, "upload-the-docker-image-to-dockerhub"]], "Usage": [[2, "usage"], [11, "usage"]], "Useful Build-Time Flags": [[80, null]], "Useful Runtime Options": [[81, null]], "Using Medusa with TensorRT-LLM": [[12, "using-medusa-with-tensorrt-llm"]], "Validated Networks for Benchmarking": [[73, "validated-networks-for-benchmarking"]], "Variables": [[74, "variables"]], "Visualize the PyTorch profiler results": [[72, "visualize-the-pytorch-profiler-results"]], "WIP: Chunked context support on DeepSeek models": [[20, "wip-chunked-context-support-on-deepseek-models"]], "WIP: Enable more features by default": [[20, "wip-enable-more-features-by-default"]], "Weight Bindings": [[16, "weight-bindings"]], "Weight Loading": [[95, "weight-loading"]], "Weights absorb and MQA": [[28, "weights-absorb-and-mqa"]], "Welcome to TensorRT-LLM\u2019s Documentation!": [[64, null]], "What Can You Do With TensorRT-LLM?": [[71, "what-can-you-do-with-tensorrt-llm"]], "What Triggers an Event?": [[8, "what-triggers-an-event"]], "What is H100 FP8?": [[22, "what-is-h100-fp8"]], "What\u2019s coming next": [[25, "whats-coming-next"]], "When to Use Graph Rewriting?": [[7, "when-to-use-graph-rewriting"]], "WindowBlockManager/BlockManager": [[8, "windowblockmanager-blockmanager"]], "Workflow": [[17, "workflow"], [73, "workflow"]], "Workload Profile": [[26, "workload-profile"]], "World Configuration": [[6, "world-configuration"]], "XQA Optimization": [[5, "xqa-optimization"]], "bufferManager.h": [[1, "buffermanager-h"]], "cacheCommunicator.h": [[0, "cachecommunicator-h"]], "common.h": [[1, "common-h"]], "cudaEvent.h": [[1, "cudaevent-h"]], "cudaStream.h": [[1, "cudastream-h"]], "dataTransceiverState.h": [[0, "datatransceiverstate-h"]], "decoderState.h": [[1, "decoderstate-h"]], "decodingInput.h": [[1, "decodinginput-h"]], "decodingOutput.h": [[1, "decodingoutput-h"]], "disaggServerUtil.h": [[0, "disaggserverutil-h"]], "disaggregated": [[30, "trtllm-serve-disaggregated"]], "disaggregated_mpi_worker": [[30, "trtllm-serve-disaggregated-mpi-worker"]], "eagleBuffers.h": [[1, "eaglebuffers-h"]], "eagleModule.h": [[1, "eaglemodule-h"]], "executor.h": [[0, "executor-h"]], "explicitDraftTokensBuffers.h": [[1, "explicitdrafttokensbuffers-h"]], "gptDecoder.h": [[1, "gptdecoder-h"]], "gptDecoderBatched.h": [[1, "gptdecoderbatched-h"]], "gptJsonConfig.h": [[1, "gptjsonconfig-h"]], "iBuffer.h": [[1, "ibuffer-h"]], "iGptDecoderBatched.h": [[1, "igptdecoderbatched-h"]], "iTensor.h": [[1, "itensor-h"]], "ipcNvlsMemory.h": [[1, "ipcnvlsmemory-h"]], "ipcUtils.h": [[1, "ipcutils-h"]], "lookaheadBuffers.h": [[1, "lookaheadbuffers-h"]], "lookaheadModule.h": [[1, "lookaheadmodule-h"]], "loraCache.h": [[1, "loracache-h"]], "loraCachePageManagerConfig.h": [[1, "loracachepagemanagerconfig-h"]], "loraModule.h": [[1, "loramodule-h"]], "medusaModule.h": [[1, "medusamodule-h"]], "memoryCounters.h": [[1, "memorycounters-h"]], "modelConfig.h": [[1, "modelconfig-h"]], "promptTuningParams.h": [[1, "prompttuningparams-h"]], "rawEngine.h": [[1, "rawengine-h"]], "request.h": [[1, "request-h"]], "runtimeDefaults.h": [[1, "runtimedefaults-h"]], "samplingConfig.h": [[1, "samplingconfig-h"]], "serialization.h": [[0, "serialization-h"]], "serve": [[30, "trtllm-serve-serve"]], "speculativeDecodingMode.h": [[1, "speculativedecodingmode-h"]], "speculativeDecodingModule.h": [[1, "speculativedecodingmodule-h"]], "tensor.h": [[0, "tensor-h"]], "tllmLogger.h": [[1, "tllmlogger-h"]], "transferAgent.h": [[0, "transferagent-h"]], "trtllm-build": [[29, null]], "trtllm-serve": [[30, null], [30, "trtllm-serve"]], "types.h": [[0, "types-h"]], "worldConfig.h": [[1, "worldconfig-h"]]}, "docnames": ["_cpp_gen/executor", "_cpp_gen/runtime", "advanced/disaggregated-service", "advanced/executor", "advanced/expert-parallelism", "advanced/gpt-attention", "advanced/gpt-runtime", "advanced/graph-rewriting", "advanced/kv-cache-management", "advanced/kv-cache-reuse", "advanced/lora", "advanced/lowprecision-pcie-allreduce", "advanced/speculative-decoding", "advanced/weight-streaming", "architecture/add-model", "architecture/checkpoint", "architecture/core-concepts", "architecture/model-weights-loader", "architecture/overview", "architecture/workflow", "blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM", "blogs/Falcon180B-H200", "blogs/H100vsA100", "blogs/H200launch", "blogs/XQA-kernel", "blogs/quantization-in-TRT-LLM", "blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs", "blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization", "blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs", "commands/trtllm-build", "commands/trtllm-serve", "dev-on-cloud/build-image-to-dockerhub", "dev-on-cloud/dev-on-runpod", "examples/curl_chat_client", "examples/curl_chat_client_for_multimodal", "examples/curl_completion_client", "examples/customization", "examples/deepseek_r1_reasoning_parser", "examples/genai_perf_client", "examples/genai_perf_client_for_multimodal", "examples/index", "examples/llm_api_examples", "examples/llm_auto_parallel", "examples/llm_eagle2_decoding", "examples/llm_eagle_decoding", "examples/llm_guided_decoding", "examples/llm_inference", "examples/llm_inference_async", "examples/llm_inference_async_streaming", "examples/llm_inference_customize", "examples/llm_inference_distributed", "examples/llm_inference_kv_events", "examples/llm_logits_processor", "examples/llm_lookahead_decoding", "examples/llm_medusa_decoding", "examples/llm_mgmn_llm_distributed", "examples/llm_mgmn_trtllm_bench", "examples/llm_mgmn_trtllm_serve", "examples/llm_multilora", "examples/llm_quantization", "examples/openai_chat_client", "examples/openai_chat_client_for_multimodal", "examples/openai_completion_client", "examples/trtllm_serve_examples", "index", "installation/build-from-source-linux", "installation/grace-hopper", "installation/linux", "key-features", "llm-api/index", "llm-api/reference", "overview", "performance/perf-analysis", "performance/perf-benchmarking", "performance/perf-overview", "performance/performance-tuning-guide/benchmarking-default-performance", "performance/performance-tuning-guide/deciding-model-sharding-strategy", "performance/performance-tuning-guide/fp8-quantization", "performance/performance-tuning-guide/index", "performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens", "performance/performance-tuning-guide/useful-build-time-flags", "performance/performance-tuning-guide/useful-runtime-flags", "python-api/tensorrt_llm.functional", "python-api/tensorrt_llm.layers", "python-api/tensorrt_llm.models", "python-api/tensorrt_llm.plugin", "python-api/tensorrt_llm.quantization", "python-api/tensorrt_llm.runtime", "quick-start-guide", "reference/memory", "reference/precision", "reference/support-matrix", "reference/troubleshooting", "release-notes", "torch", "torch/adding_new_model", "torch/arch_overview", "torch/attention", "torch/kv_cache_manager", "torch/scheduler"], "envversion": {"sphinx": 62, "sphinx.domains.c": 3, "sphinx.domains.changeset": 1, "sphinx.domains.citation": 1, "sphinx.domains.cpp": 9, "sphinx.domains.index": 1, "sphinx.domains.javascript": 3, "sphinx.domains.math": 2, "sphinx.domains.python": 4, "sphinx.domains.rst": 2, "sphinx.domains.std": 2, "sphinx.ext.todo": 2, "sphinx.ext.viewcode": 1}, "filenames": ["_cpp_gen/executor.rst", "_cpp_gen/runtime.rst", "advanced/disaggregated-service.md", "advanced/executor.md", "advanced/expert-parallelism.md", "advanced/gpt-attention.md", "advanced/gpt-runtime.md", "advanced/graph-rewriting.md", "advanced/kv-cache-management.md", "advanced/kv-cache-reuse.md", "advanced/lora.md", "advanced/lowprecision-pcie-allreduce.md", "advanced/speculative-decoding.md", "advanced/weight-streaming.md", "architecture/add-model.md", "architecture/checkpoint.md", "architecture/core-concepts.md", "architecture/model-weights-loader.md", "architecture/overview.md", "architecture/workflow.md", "blogs/Best_perf_practice_on_DeepSeek-R1_in_TensorRT-LLM.md", "blogs/Falcon180B-H200.md", "blogs/H100vsA100.md", "blogs/H200launch.md", "blogs/XQA-kernel.md", "blogs/quantization-in-TRT-LLM.md", "blogs/tech_blog/blog1_Pushing_Latency_Boundaries_Optimizing_DeepSeek-R1_Performance_on_NVIDIA_B200_GPUs.md", "blogs/tech_blog/blog2_DeepSeek_R1_MTP_Implementation_and_Optimization.md", "blogs/tech_blog/blog3_Optimizing_DeepSeek_R1_Throughput_on_NVIDIA_Blackwell_GPUs.md", "commands/trtllm-build.rst", "commands/trtllm-serve.rst", "dev-on-cloud/build-image-to-dockerhub.md", "dev-on-cloud/dev-on-runpod.md", "examples/curl_chat_client.rst", "examples/curl_chat_client_for_multimodal.rst", "examples/curl_completion_client.rst", "examples/customization.md", "examples/deepseek_r1_reasoning_parser.rst", "examples/genai_perf_client.rst", "examples/genai_perf_client_for_multimodal.rst", "examples/index.rst", "examples/llm_api_examples.rst", "examples/llm_auto_parallel.rst", "examples/llm_eagle2_decoding.rst", "examples/llm_eagle_decoding.rst", "examples/llm_guided_decoding.rst", "examples/llm_inference.rst", "examples/llm_inference_async.rst", "examples/llm_inference_async_streaming.rst", "examples/llm_inference_customize.rst", "examples/llm_inference_distributed.rst", "examples/llm_inference_kv_events.rst", "examples/llm_logits_processor.rst", "examples/llm_lookahead_decoding.rst", "examples/llm_medusa_decoding.rst", "examples/llm_mgmn_llm_distributed.rst", "examples/llm_mgmn_trtllm_bench.rst", "examples/llm_mgmn_trtllm_serve.rst", "examples/llm_multilora.rst", "examples/llm_quantization.rst", "examples/openai_chat_client.rst", "examples/openai_chat_client_for_multimodal.rst", "examples/openai_completion_client.rst", "examples/trtllm_serve_examples.rst", "index.rst", "installation/build-from-source-linux.md", "installation/grace-hopper.md", "installation/linux.md", "key-features.md", "llm-api/index.md", "llm-api/reference.rst", "overview.md", "performance/perf-analysis.md", "performance/perf-benchmarking.md", "performance/perf-overview.md", "performance/performance-tuning-guide/benchmarking-default-performance.md", "performance/performance-tuning-guide/deciding-model-sharding-strategy.md", "performance/performance-tuning-guide/fp8-quantization.md", "performance/performance-tuning-guide/index.rst", "performance/performance-tuning-guide/tuning-max-batch-size-and-max-num-tokens.md", "performance/performance-tuning-guide/useful-build-time-flags.md", "performance/performance-tuning-guide/useful-runtime-flags.md", "python-api/tensorrt_llm.functional.rst", "python-api/tensorrt_llm.layers.rst", "python-api/tensorrt_llm.models.rst", "python-api/tensorrt_llm.plugin.rst", "python-api/tensorrt_llm.quantization.rst", "python-api/tensorrt_llm.runtime.rst", "quick-start-guide.md", "reference/memory.md", "reference/precision.md", "reference/support-matrix.md", "reference/troubleshooting.md", "release-notes.md", "torch.md", "torch/adding_new_model.md", "torch/arch_overview.md", "torch/attention.md", "torch/kv_cache_manager.md", "torch/scheduler.md"], "indexentries": {"--backend": [[30, "cmdoption-trtllm-serve-serve-backend", false]], "--cluster_size": [[30, "cmdoption-trtllm-serve-serve-cluster_size", false]], "--config_file": [[30, "cmdoption-trtllm-serve-disaggregated-c", false], [30, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false]], "--ep_size": [[30, "cmdoption-trtllm-serve-serve-ep_size", false]], "--extra_llm_api_options": [[30, "cmdoption-trtllm-serve-serve-extra_llm_api_options", false]], "--gpus_per_node": [[30, "cmdoption-trtllm-serve-serve-gpus_per_node", false]], "--host": [[30, "cmdoption-trtllm-serve-serve-host", false]], "--kv_cache_free_gpu_memory_fraction": [[30, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", false]], "--log_level": [[30, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", false], [30, "cmdoption-trtllm-serve-serve-log_level", false]], "--max_batch_size": [[30, "cmdoption-trtllm-serve-serve-max_batch_size", false]], "--max_beam_width": [[30, "cmdoption-trtllm-serve-serve-max_beam_width", false]], "--max_num_tokens": [[30, "cmdoption-trtllm-serve-serve-max_num_tokens", false]], "--max_seq_len": [[30, "cmdoption-trtllm-serve-serve-max_seq_len", false]], "--num_postprocess_workers": [[30, "cmdoption-trtllm-serve-serve-num_postprocess_workers", false]], "--port": [[30, "cmdoption-trtllm-serve-serve-port", false]], "--pp_size": [[30, "cmdoption-trtllm-serve-serve-pp_size", false]], "--reasoning_parser": [[30, "cmdoption-trtllm-serve-serve-reasoning_parser", false]], "--request_timeout": [[30, "cmdoption-trtllm-serve-disaggregated-r", false]], "--server_start_timeout": [[30, "cmdoption-trtllm-serve-disaggregated-t", false]], "--tokenizer": [[30, "cmdoption-trtllm-serve-serve-tokenizer", false]], "--tp_size": [[30, "cmdoption-trtllm-serve-serve-tp_size", false]], "--trust_remote_code": [[30, "cmdoption-trtllm-serve-serve-trust_remote_code", false]], "-c": [[30, "cmdoption-trtllm-serve-disaggregated-c", false], [30, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false]], "-r": [[30, "cmdoption-trtllm-serve-disaggregated-r", false]], "-t": [[30, "cmdoption-trtllm-serve-disaggregated-t", false]], "__init__() (tensorrt_llm.llmapi.buildcacheconfig method)": [[70, "tensorrt_llm.llmapi.BuildCacheConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.buildconfig method)": [[70, "tensorrt_llm.llmapi.BuildConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.completionoutput method)": [[70, "tensorrt_llm.llmapi.CompletionOutput.__init__", false]], "__init__() (tensorrt_llm.llmapi.disaggregatedparams method)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.__init__", false]], "__init__() (tensorrt_llm.llmapi.guideddecodingparams method)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.__init__", false]], "__init__() (tensorrt_llm.llmapi.kvcacheretentionconfig method)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig method)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.__init__", false]], "__init__() (tensorrt_llm.llmapi.lookaheaddecodingconfig method)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.__init__", false]], "__init__() (tensorrt_llm.llmapi.quantconfig method)": [[70, "tensorrt_llm.llmapi.QuantConfig.__init__", false]], "__init__() (tensorrt_llm.llmapi.requestoutput method)": [[70, "tensorrt_llm.llmapi.RequestOutput.__init__", false]], "__init__() (tensorrt_llm.llmapi.samplingparams method)": [[70, "tensorrt_llm.llmapi.SamplingParams.__init__", false]], "abort() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.abort", false]], "abs() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.abs", false]], "abs() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.abs", false]], "activation() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.activation", false]], "adalayernorm (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNorm", false]], "adalayernormcontinuous (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormContinuous", false]], "adalayernormzero (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormZero", false]], "adalayernormzerosingle (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle", false]], "add() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.add", false]], "add_input() (tensorrt_llm.functional.conditional method)": [[82, "tensorrt_llm.functional.Conditional.add_input", false]], "add_output() (tensorrt_llm.functional.conditional method)": [[82, "tensorrt_llm.functional.Conditional.add_output", false]], "add_sequence() (tensorrt_llm.runtime.kvcachemanager method)": [[87, "tensorrt_llm.runtime.KVCacheManager.add_sequence", false]], "add_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.add_special_tokens", false]], "additional_model_outputs (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.additional_model_outputs", false]], "alibi (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.alibi", false]], "alibi_with_scale (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.alibi_with_scale", false]], "allgather() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.allgather", false]], "allreduce() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.allreduce", false]], "allreducefusionop (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.AllReduceFusionOp", false]], "allreduceparams (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.AllReduceParams", false]], "allreducestrategy (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.AllReduceStrategy", false]], "apply_batched_logits_processor (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.apply_batched_logits_processor", false]], "apply_llama3_scaling() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_llama3_scaling", false]], "apply_rotary_pos_emb() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb", false]], "apply_rotary_pos_emb_chatglm() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb_chatglm", false]], "apply_rotary_pos_emb_cogvlm() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.apply_rotary_pos_emb_cogvlm", false]], "arange() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.arange", false]], "argmax() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.argmax", false]], "assert_valid_quant_algo() (tensorrt_llm.models.gemmaforcausallm class method)": [[84, "tensorrt_llm.models.GemmaForCausalLM.assert_valid_quant_algo", false]], "assertion() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.assertion", false]], "attention (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.Attention", false]], "attentionmaskparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.AttentionMaskParams", false]], "attentionmasktype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.AttentionMaskType", false]], "attentionparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.AttentionParams", false]], "attn_backend (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.attn_backend", false]], "attn_processors (tensorrt_llm.models.sd3transformer2dmodel property)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.attn_processors", false]], "audio_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.audio_engine_dir", false]], "auto (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.AUTO", false]], "auto_deploy_config (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.auto_deploy_config", false]], "auto_parallel (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.auto_parallel", false]], "auto_parallel_config (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.auto_parallel_config", false]], "auto_parallel_config (tensorrt_llm.llmapi.trtllmargs property)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.auto_parallel_config", false]], "auto_parallel_world_size (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.auto_parallel_world_size", false]], "autotuner_enabled (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.autotuner_enabled", false]], "avg_pool2d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.avg_pool2d", false]], "avgpool2d (class in tensorrt_llm.layers.pooling)": [[83, "tensorrt_llm.layers.pooling.AvgPool2d", false]], "axes (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.axes", false]], "bad (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.bad", false]], "bad_token_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.bad_token_ids", false]], "bad_words_list (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.bad_words_list", false]], "baichuanforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BaichuanForCausalLM", false]], "batch_size (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.batch_size", false]], "batchingtype (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.BatchingType", false]], "beam_search_diversity_rate (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.beam_search_diversity_rate", false]], "beam_search_diversity_rate (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.beam_search_diversity_rate", false]], "beam_width_array (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.beam_width_array", false]], "bert_attention() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.bert_attention", false]], "bertattention (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.BertAttention", false]], "bertforquestionanswering (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BertForQuestionAnswering", false]], "bertforsequenceclassification (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BertForSequenceClassification", false]], "bertmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BertModel", false]], "best_of (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.best_of", false]], "bidirectional (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.bidirectional", false]], "bidirectionalglm (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.bidirectionalglm", false]], "blocksparse (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.blocksparse", false]], "blocksparseattnparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.BlockSparseAttnParams", false]], "bloomforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BloomForCausalLM", false]], "bloommodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.BloomModel", false]], "broadcast_helper() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.broadcast_helper", false]], "buffer_allocated (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.buffer_allocated", false]], "build_config (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.build_config", false]], "build_config (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.build_config", false]], "buildcacheconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.BuildCacheConfig", false]], "buildconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.BuildConfig", false]], "cache_root (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildCacheConfig.cache_root", false]], "cache_root (tensorrt_llm.llmapi.buildcacheconfig property)": [[70, "id7", false]], "cachetransceiverconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.CacheTransceiverConfig", false]], "calculate_speculative_resource() (tensorrt_llm.llmapi.lookaheaddecodingconfig method)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.calculate_speculative_resource", false]], "calib_batch_size (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.calib_batch_size", false]], "calib_batches (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.calib_batches", false]], "calib_config (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.calib_config", false]], "calib_dataset (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.calib_dataset", false]], "calib_max_seq_length (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.calib_max_seq_length", false]], "calibconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.CalibConfig", false]], "capacity_scheduler_policy (tensorrt_llm.llmapi.schedulerconfig attribute)": [[70, "tensorrt_llm.llmapi.SchedulerConfig.capacity_scheduler_policy", false]], "capacityschedulerpolicy (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.CapacitySchedulerPolicy", false]], "cast (class in tensorrt_llm.layers.cast)": [[83, "tensorrt_llm.layers.cast.Cast", false]], "cast() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.cast", false]], "cast() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.cast", false]], "categorical_sample() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.categorical_sample", false]], "causal (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.causal", false]], "chatglm (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.chatglm", false]], "chatglmconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.ChatGLMConfig", false]], "chatglmforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.ChatGLMForCausalLM", false]], "chatglmgenerationsession (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.ChatGLMGenerationSession", false]], "chatglmmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.ChatGLMModel", false]], "check_config() (tensorrt_llm.models.decodermodel method)": [[84, "tensorrt_llm.models.DecoderModel.check_config", false]], "check_config() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.check_config", false]], "check_config() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.check_config", false]], "check_config() (tensorrt_llm.models.falconforcausallm method)": [[84, "tensorrt_llm.models.FalconForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.mptforcausallm method)": [[84, "tensorrt_llm.models.MPTForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.optforcausallm method)": [[84, "tensorrt_llm.models.OPTForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.phiforcausallm method)": [[84, "tensorrt_llm.models.PhiForCausalLM.check_config", false]], "check_config() (tensorrt_llm.models.pretrainedmodel method)": [[84, "tensorrt_llm.models.PretrainedModel.check_config", false]], "choices() (tensorrt_llm.functional.positionembeddingtype static method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.choices", false]], "chunk() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.chunk", false]], "clamp_val (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.clamp_val", false]], "clip() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.clip", false]], "clipvisiontransformer (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.CLIPVisionTransformer", false]], "cogvlmattention (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.CogVLMAttention", false]], "cogvlmconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.CogVLMConfig", false]], "cogvlmforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.CogVLMForCausalLM", false]], "cohereforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.CohereForCausalLM", false]], "collect_and_bias() (tensorrt_llm.layers.linear.linear method)": [[83, "tensorrt_llm.layers.linear.Linear.collect_and_bias", false]], "collect_and_bias() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.collect_and_bias", false]], "collect_and_bias() (tensorrt_llm.layers.linear.rowlinear method)": [[83, "tensorrt_llm.layers.linear.RowLinear.collect_and_bias", false]], "columnlinear (in module tensorrt_llm.layers.linear)": [[83, "tensorrt_llm.layers.linear.ColumnLinear", false]], "combinedtimesteplabelembeddings (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings", false]], "combinedtimesteptextprojembeddings (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings", false]], "completionoutput (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.CompletionOutput", false]], "compute_relative_bias() (in module tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.compute_relative_bias", false]], "concat() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.concat", false]], "conditional (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.Conditional", false]], "config_class (tensorrt_llm.models.baichuanforcausallm attribute)": [[84, "tensorrt_llm.models.BaichuanForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.chatglmforcausallm attribute)": [[84, "tensorrt_llm.models.ChatGLMForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.cogvlmforcausallm attribute)": [[84, "tensorrt_llm.models.CogVLMForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.cohereforcausallm attribute)": [[84, "tensorrt_llm.models.CohereForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.dbrxforcausallm attribute)": [[84, "tensorrt_llm.models.DbrxForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.deepseekforcausallm attribute)": [[84, "tensorrt_llm.models.DeepseekForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.deepseekv2forcausallm attribute)": [[84, "tensorrt_llm.models.DeepseekV2ForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.eagleforcausallm attribute)": [[84, "tensorrt_llm.models.EagleForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.falconforcausallm attribute)": [[84, "tensorrt_llm.models.FalconForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gemmaforcausallm attribute)": [[84, "tensorrt_llm.models.GemmaForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gptforcausallm attribute)": [[84, "tensorrt_llm.models.GPTForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.gptjforcausallm attribute)": [[84, "tensorrt_llm.models.GPTJForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.llamaforcausallm attribute)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.mambaforcausallm attribute)": [[84, "tensorrt_llm.models.MambaForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.medusaforcausallm attribute)": [[84, "tensorrt_llm.models.MedusaForCausalLm.config_class", false]], "config_class (tensorrt_llm.models.mllamaforcausallm attribute)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.phi3forcausallm attribute)": [[84, "tensorrt_llm.models.Phi3ForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.phiforcausallm attribute)": [[84, "tensorrt_llm.models.PhiForCausalLM.config_class", false]], "config_class (tensorrt_llm.models.sd3transformer2dmodel attribute)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.config_class", false]], "constant() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.constant", false]], "constant_to_tensor_() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.constant_to_tensor_", false]], "constants_to_tensors_() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.constants_to_tensors_", false]], "context (tensorrt_llm.runtime.session property)": [[87, "tensorrt_llm.runtime.Session.context", false]], "context_chunking_policy (tensorrt_llm.llmapi.schedulerconfig attribute)": [[70, "tensorrt_llm.llmapi.SchedulerConfig.context_chunking_policy", false]], "context_logits (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.context_logits", false]], "context_mem_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.context_mem_size", false]], "context_mem_size (tensorrt_llm.runtime.session property)": [[87, "tensorrt_llm.runtime.Session.context_mem_size", false]], "contextchunkingpolicy (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.ContextChunkingPolicy", false]], "conv1d (class in tensorrt_llm.layers.conv)": [[83, "tensorrt_llm.layers.conv.Conv1d", false]], "conv1d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.conv1d", false]], "conv2d (class in tensorrt_llm.layers.conv)": [[83, "tensorrt_llm.layers.conv.Conv2d", false]], "conv2d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.conv2d", false]], "conv3d (class in tensorrt_llm.layers.conv)": [[83, "tensorrt_llm.layers.conv.Conv3d", false]], "conv3d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.conv3d", false]], "conv_kernel (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.conv_kernel", false]], "conv_kernel (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.conv_kernel", false]], "conv_transpose2d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.conv_transpose2d", false]], "convert_load_format() (tensorrt_llm.llmapi.torchllmargs class method)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.convert_load_format", false]], "convtranspose2d (class in tensorrt_llm.layers.conv)": [[83, "tensorrt_llm.layers.conv.ConvTranspose2d", false]], "copy_on_partial_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.copy_on_partial_reuse", false]], "cos() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.cos", false]], "cp_split_plugin() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.cp_split_plugin", false]], "cpp_e2e (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.cpp_e2e", false]], "cpp_llm_only (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.cpp_llm_only", false]], "create_allreduce_plugin() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.create_allreduce_plugin", false]], "create_attention_const_params() (tensorrt_llm.layers.attention.attention static method)": [[83, "tensorrt_llm.layers.attention.Attention.create_attention_const_params", false]], "create_fake_weight() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_fake_weight", false]], "create_runtime_defaults() (tensorrt_llm.models.pretrainedconfig static method)": [[84, "tensorrt_llm.models.PretrainedConfig.create_runtime_defaults", false]], "create_sinusoidal_positions() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions", false]], "create_sinusoidal_positions_for_attention_plugin() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_for_attention_plugin", false]], "create_sinusoidal_positions_for_cogvlm_attention_plugin() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_for_cogvlm_attention_plugin", false]], "create_sinusoidal_positions_long_rope() (tensorrt_llm.functional.ropeembeddingutils method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_long_rope", false]], "create_sinusoidal_positions_yarn() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.create_sinusoidal_positions_yarn", false]], "cropped_pos_embed() (tensorrt_llm.layers.embedding.sd3patchembed method)": [[83, "tensorrt_llm.layers.embedding.SD3PatchEmbed.cropped_pos_embed", false]], "cross_attention (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.cross_attention", false]], "cross_attention (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.cross_attention", false]], "cross_kv_cache_fraction (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.cross_kv_cache_fraction", false]], "ctx_request_id (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.ctx_request_id", false]], "cuda_graph_batch_sizes (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.cuda_graph_batch_sizes", false]], "cuda_graph_cache_size (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.cuda_graph_cache_size", false]], "cuda_graph_max_batch_size (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.cuda_graph_max_batch_size", false]], "cuda_graph_mode (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.cuda_graph_mode", false]], "cuda_graph_mode (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.cuda_graph_mode", false]], "cuda_graph_padding_enabled (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.cuda_graph_padding_enabled", false]], "cuda_stream_guard() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.cuda_stream_guard", false]], "cuda_stream_sync() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.cuda_stream_sync", false]], "cumsum() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.cumsum", false]], "cumulative_logprob (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.cumulative_logprob", false]], "custom_mask (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.custom_mask", false]], "data (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.data", false]], "dbrxconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DbrxConfig", false]], "dbrxforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DbrxForCausalLM", false]], "debug_mode (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.debug_mode", false]], "debug_tensors_to_save (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.debug_tensors_to_save", false]], "decode() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.decode", false]], "decode_batch() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.decode_batch", false]], "decode_duration_ms (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.decode_duration_ms", false]], "decode_regular() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.decode_regular", false]], "decode_retention_priority (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.decode_retention_priority", false]], "decode_stream() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.decode_stream", false]], "decode_words_list() (in module tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.decode_words_list", false]], "decodermodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DecoderModel", false]], "decoding_config (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.decoding_config", false]], "decoding_config (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.decoding_config", false]], "decoding_type (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.decoding_type", false]], "decoding_type (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.decoding_type", false]], "deepseekforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DeepseekForCausalLM", false]], "deepseekv2attention (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.DeepseekV2Attention", false]], "deepseekv2forcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DeepseekV2ForCausalLM", false]], "default_plugin_config() (tensorrt_llm.models.cogvlmforcausallm method)": [[84, "tensorrt_llm.models.CogVLMForCausalLM.default_plugin_config", false]], "default_plugin_config() (tensorrt_llm.models.llamaforcausallm method)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.default_plugin_config", false]], "deferred (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.deferred", false]], "detokenize (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.detokenize", false]], "device (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.device", false]], "device (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.device", false]], "diffusersattention (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.DiffusersAttention", false]], "dimrange (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.DimRange", false]], "directory (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.directory", false]], "disable (tensorrt_llm.functional.sidestreamidtype attribute)": [[82, "tensorrt_llm.functional.SideStreamIDType.disable", false]], "disable_forward_chunking() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.disable_forward_chunking", false]], "disable_overlap_scheduler (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.disable_overlap_scheduler", false]], "disaggregated_params (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.disaggregated_params", false]], "disaggregatedparams (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams", false]], "dit (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.DiT", false]], "div() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.div", false]], "dora_plugin() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.dora_plugin", false]], "draft_tokens (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.draft_tokens", false]], "draft_tokens_external (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.DRAFT_TOKENS_EXTERNAL", false]], "dry_run (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.dry_run", false]], "dtype (tensorrt_llm.functional.tensor property)": [[82, "tensorrt_llm.functional.Tensor.dtype", false]], "dtype (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.dtype", false]], "dtype (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.dtype", false]], "dtype (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.dtype", false]], "dtype (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.dtype", false]], "dtype (tensorrt_llm.runtime.tensorinfo attribute)": [[87, "tensorrt_llm.runtime.TensorInfo.dtype", false]], "dump_debug_buffers() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.dump_debug_buffers", false]], "duration_ms (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.duration_ms", false]], "dynamic (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.dynamic", false]], "dynamic_batch_config (tensorrt_llm.llmapi.schedulerconfig attribute)": [[70, "tensorrt_llm.llmapi.SchedulerConfig.dynamic_batch_config", false]], "dynamic_batch_moving_average_window (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[70, "tensorrt_llm.llmapi.DynamicBatchConfig.dynamic_batch_moving_average_window", false]], "dynamic_tree_max_topk (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.dynamic_tree_max_topK", false]], "dynamicbatchconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.DynamicBatchConfig", false]], "eagle (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.EAGLE", false]], "eagle3_one_model (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.eagle3_one_model", false]], "eagle_choices (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.eagle_choices", false]], "eagledecodingconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig", false]], "eagleforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.EagleForCausalLM", false]], "early_stop_criteria() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.early_stop_criteria", false]], "early_stopping (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.early_stopping", false]], "early_stopping (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.early_stopping", false]], "einsum() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.einsum", false]], "elementwise_binary() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.elementwise_binary", false]], "embedding (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.Embedding", false]], "embedding() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.embedding", false]], "embedding_bias (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.embedding_bias", false]], "embedding_parallel_mode (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.embedding_parallel_mode", false]], "enable_batch_size_tuning (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[70, "tensorrt_llm.llmapi.DynamicBatchConfig.enable_batch_size_tuning", false]], "enable_block_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.enable_block_reuse", false]], "enable_build_cache (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.enable_build_cache", false]], "enable_context_fmha_fp32_acc (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.enable_context_fmha_fp32_acc", false]], "enable_debug_output (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.enable_debug_output", false]], "enable_forward_chunking() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.enable_forward_chunking", false]], "enable_iter_perf_stats (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.enable_iter_perf_stats", false]], "enable_iter_req_stats (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.enable_iter_req_stats", false]], "enable_layerwise_nvtx_marker (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.enable_layerwise_nvtx_marker", false]], "enable_max_num_tokens_tuning (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[70, "tensorrt_llm.llmapi.DynamicBatchConfig.enable_max_num_tokens_tuning", false]], "enable_min_latency (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.enable_min_latency", false]], "enable_partial_reuse (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.enable_partial_reuse", false]], "enable_tqdm (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.enable_tqdm", false]], "enable_trtllm_sampler (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.enable_trtllm_sampler", false]], "encdecmodelrunner (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.EncDecModelRunner", false]], "encoder_run() (tensorrt_llm.runtime.encdecmodelrunner method)": [[87, "tensorrt_llm.runtime.EncDecModelRunner.encoder_run", false]], "encodermodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.EncoderModel", false]], "end_id (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.end_id", false]], "end_id (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.end_id", false]], "engine (tensorrt_llm.runtime.session property)": [[87, "tensorrt_llm.runtime.Session.engine", false]], "engine_inspector (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.engine_inspector", false]], "eq() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.eq", false]], "equal_progress (tensorrt_llm.llmapi.contextchunkingpolicy attribute)": [[70, "tensorrt_llm.llmapi.ContextChunkingPolicy.EQUAL_PROGRESS", false]], "event_buffer_max_size (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.event_buffer_max_size", false]], "exclude_input_from_output (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.exclude_input_from_output", false]], "exclude_modules (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.exclude_modules", false]], "exp() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.exp", false]], "expand() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.expand", false]], "expand_dims() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.expand_dims", false]], "expand_dims_like() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.expand_dims_like", false]], "expand_mask() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.expand_mask", false]], "explicit_draft_tokens (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.EXPLICIT_DRAFT_TOKENS", false]], "extended_runtime_perf_knob_config (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.extended_runtime_perf_knob_config", false]], "extendedruntimeperfknobconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig", false]], "extra_resource_managers (tensorrt_llm.llmapi.torchllmargs property)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.extra_resource_managers", false]], "falconconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.FalconConfig", false]], "falconforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.FalconForCausalLM", false]], "falconmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.FalconModel", false]], "fast_build (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.fast_build", false]], "fc_gate() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[83, "tensorrt_llm.layers.mlp.FusedGatedMLP.fc_gate", false]], "fc_gate_dora() (in module tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.fc_gate_dora", false]], "fc_gate_lora() (in module tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.fc_gate_lora", false]], "fc_gate_plugin() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[83, "tensorrt_llm.layers.mlp.FusedGatedMLP.fc_gate_plugin", false]], "field_name (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "id12", false], [70, "id15", false], [70, "id18", false], [70, "tensorrt_llm.llmapi.TorchLlmArgs.field_name", false]], "field_name (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "id21", false], [70, "id24", false], [70, "id27", false], [70, "id30", false], [70, "id33", false], [70, "tensorrt_llm.llmapi.TrtLlmArgs.field_name", false]], "fill_attention_const_params_for_long_rope() (tensorrt_llm.layers.attention.attentionparams method)": [[83, "tensorrt_llm.layers.attention.AttentionParams.fill_attention_const_params_for_long_rope", false]], "fill_attention_const_params_for_rope() (tensorrt_llm.layers.attention.attentionparams method)": [[83, "tensorrt_llm.layers.attention.AttentionParams.fill_attention_const_params_for_rope", false]], "fill_attention_params() (tensorrt_llm.layers.attention.attention static method)": [[83, "tensorrt_llm.layers.attention.Attention.fill_attention_params", false]], "fill_none_tensor_list() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[83, "tensorrt_llm.layers.attention.KeyValueCacheParams.fill_none_tensor_list", false]], "fill_value (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.fill_value", false]], "filter_medusa_logits() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.filter_medusa_logits", false]], "finalize_decoder() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.finalize_decoder", false]], "find_best_medusa_path() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.find_best_medusa_path", false]], "finish_reason (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.finish_reason", false]], "finished (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.finished", false]], "first_come_first_served (tensorrt_llm.llmapi.contextchunkingpolicy attribute)": [[70, "tensorrt_llm.llmapi.ContextChunkingPolicy.FIRST_COME_FIRST_SERVED", false]], "first_gen_tokens (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.first_gen_tokens", false]], "first_layer (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.first_layer", false]], "flatten() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.flatten", false]], "flatten() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.flatten", false]], "flip() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.flip", false]], "floordiv() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.floordiv", false]], "fmt_dim (c macro)": [[1, "c.FMT_DIM", false]], "for_each_rank() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.for_each_rank", false]], "force_num_profiles (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.force_num_profiles", false]], "forward() (tensorrt_llm.layers.activation.mish method)": [[83, "tensorrt_llm.layers.activation.Mish.forward", false]], "forward() (tensorrt_llm.layers.attention.attention method)": [[83, "tensorrt_llm.layers.attention.Attention.forward", false]], "forward() (tensorrt_llm.layers.attention.bertattention method)": [[83, "tensorrt_llm.layers.attention.BertAttention.forward", false]], "forward() (tensorrt_llm.layers.attention.cogvlmattention method)": [[83, "tensorrt_llm.layers.attention.CogVLMAttention.forward", false]], "forward() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[83, "tensorrt_llm.layers.attention.DeepseekV2Attention.forward", false]], "forward() (tensorrt_llm.layers.attention.diffusersattention method)": [[83, "tensorrt_llm.layers.attention.DiffusersAttention.forward", false]], "forward() (tensorrt_llm.layers.cast.cast method)": [[83, "tensorrt_llm.layers.cast.Cast.forward", false]], "forward() (tensorrt_llm.layers.conv.conv1d method)": [[83, "tensorrt_llm.layers.conv.Conv1d.forward", false]], "forward() (tensorrt_llm.layers.conv.conv2d method)": [[83, "tensorrt_llm.layers.conv.Conv2d.forward", false]], "forward() (tensorrt_llm.layers.conv.conv3d method)": [[83, "tensorrt_llm.layers.conv.Conv3d.forward", false]], "forward() (tensorrt_llm.layers.conv.convtranspose2d method)": [[83, "tensorrt_llm.layers.conv.ConvTranspose2d.forward", false]], "forward() (tensorrt_llm.layers.embedding.combinedtimesteplabelembeddings method)": [[83, "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings.forward", false]], "forward() (tensorrt_llm.layers.embedding.combinedtimesteptextprojembeddings method)": [[83, "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings.forward", false]], "forward() (tensorrt_llm.layers.embedding.embedding method)": [[83, "tensorrt_llm.layers.embedding.Embedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.labelembedding method)": [[83, "tensorrt_llm.layers.embedding.LabelEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.pixartalphatextprojection method)": [[83, "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection.forward", false]], "forward() (tensorrt_llm.layers.embedding.prompttuningembedding method)": [[83, "tensorrt_llm.layers.embedding.PromptTuningEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.sd3patchembed method)": [[83, "tensorrt_llm.layers.embedding.SD3PatchEmbed.forward", false]], "forward() (tensorrt_llm.layers.embedding.timestepembedding method)": [[83, "tensorrt_llm.layers.embedding.TimestepEmbedding.forward", false]], "forward() (tensorrt_llm.layers.embedding.timesteps method)": [[83, "tensorrt_llm.layers.embedding.Timesteps.forward", false]], "forward() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.forward", false]], "forward() (tensorrt_llm.layers.mlp.fusedgatedmlp method)": [[83, "tensorrt_llm.layers.mlp.FusedGatedMLP.forward", false]], "forward() (tensorrt_llm.layers.mlp.gatedmlp method)": [[83, "tensorrt_llm.layers.mlp.GatedMLP.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearactivation method)": [[83, "tensorrt_llm.layers.mlp.LinearActivation.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearapproximategelu method)": [[83, "tensorrt_llm.layers.mlp.LinearApproximateGELU.forward", false]], "forward() (tensorrt_llm.layers.mlp.lineargeglu method)": [[83, "tensorrt_llm.layers.mlp.LinearGEGLU.forward", false]], "forward() (tensorrt_llm.layers.mlp.lineargelu method)": [[83, "tensorrt_llm.layers.mlp.LinearGELU.forward", false]], "forward() (tensorrt_llm.layers.mlp.linearswiglu method)": [[83, "tensorrt_llm.layers.mlp.LinearSwiGLU.forward", false]], "forward() (tensorrt_llm.layers.mlp.mlp method)": [[83, "tensorrt_llm.layers.mlp.MLP.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernorm method)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormcontinuous method)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormContinuous.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormzero method)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormZero.forward", false]], "forward() (tensorrt_llm.layers.normalization.adalayernormzerosingle method)": [[83, "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle.forward", false]], "forward() (tensorrt_llm.layers.normalization.groupnorm method)": [[83, "tensorrt_llm.layers.normalization.GroupNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.layernorm method)": [[83, "tensorrt_llm.layers.normalization.LayerNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.rmsnorm method)": [[83, "tensorrt_llm.layers.normalization.RmsNorm.forward", false]], "forward() (tensorrt_llm.layers.normalization.sd35adalayernormzerox method)": [[83, "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX.forward", false]], "forward() (tensorrt_llm.layers.pooling.avgpool2d method)": [[83, "tensorrt_llm.layers.pooling.AvgPool2d.forward", false]], "forward() (tensorrt_llm.models.bertforquestionanswering method)": [[84, "tensorrt_llm.models.BertForQuestionAnswering.forward", false]], "forward() (tensorrt_llm.models.bertforsequenceclassification method)": [[84, "tensorrt_llm.models.BertForSequenceClassification.forward", false]], "forward() (tensorrt_llm.models.bertmodel method)": [[84, "tensorrt_llm.models.BertModel.forward", false]], "forward() (tensorrt_llm.models.bloommodel method)": [[84, "tensorrt_llm.models.BloomModel.forward", false]], "forward() (tensorrt_llm.models.chatglmmodel method)": [[84, "tensorrt_llm.models.ChatGLMModel.forward", false]], "forward() (tensorrt_llm.models.clipvisiontransformer method)": [[84, "tensorrt_llm.models.CLIPVisionTransformer.forward", false]], "forward() (tensorrt_llm.models.decodermodel method)": [[84, "tensorrt_llm.models.DecoderModel.forward", false]], "forward() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.forward", false]], "forward() (tensorrt_llm.models.eagleforcausallm method)": [[84, "tensorrt_llm.models.EagleForCausalLM.forward", false]], "forward() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.forward", false]], "forward() (tensorrt_llm.models.falconmodel method)": [[84, "tensorrt_llm.models.FalconModel.forward", false]], "forward() (tensorrt_llm.models.gptjmodel method)": [[84, "tensorrt_llm.models.GPTJModel.forward", false]], "forward() (tensorrt_llm.models.gptmodel method)": [[84, "tensorrt_llm.models.GPTModel.forward", false]], "forward() (tensorrt_llm.models.gptneoxmodel method)": [[84, "tensorrt_llm.models.GPTNeoXModel.forward", false]], "forward() (tensorrt_llm.models.llamamodel method)": [[84, "tensorrt_llm.models.LLaMAModel.forward", false]], "forward() (tensorrt_llm.models.llavanextvisionwrapper method)": [[84, "tensorrt_llm.models.LlavaNextVisionWrapper.forward", false]], "forward() (tensorrt_llm.models.mambaforcausallm method)": [[84, "tensorrt_llm.models.MambaForCausalLM.forward", false]], "forward() (tensorrt_llm.models.mllamaforcausallm method)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM.forward", false]], "forward() (tensorrt_llm.models.mptmodel method)": [[84, "tensorrt_llm.models.MPTModel.forward", false]], "forward() (tensorrt_llm.models.optmodel method)": [[84, "tensorrt_llm.models.OPTModel.forward", false]], "forward() (tensorrt_llm.models.phi3model method)": [[84, "tensorrt_llm.models.Phi3Model.forward", false]], "forward() (tensorrt_llm.models.phimodel method)": [[84, "tensorrt_llm.models.PhiModel.forward", false]], "forward() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[84, "tensorrt_llm.models.RecurrentGemmaForCausalLM.forward", false]], "forward() (tensorrt_llm.models.redrafterforcausallm method)": [[84, "tensorrt_llm.models.ReDrafterForCausalLM.forward", false]], "forward() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.forward", false]], "forward() (tensorrt_llm.models.whisperencoder method)": [[84, "tensorrt_llm.models.WhisperEncoder.forward", false]], "forward_with_cfg() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.forward_with_cfg", false]], "forward_without_cfg() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.forward_without_cfg", false]], "fp8 (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.FP8", false]], "fp8_block_scales (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.FP8_BLOCK_SCALES", false]], "fp8_per_channel_per_token (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.FP8_PER_CHANNEL_PER_TOKEN", false]], "free_gpu_memory_fraction (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.free_gpu_memory_fraction", false]], "frequency_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.frequency_penalty", false]], "frequency_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.frequency_penalty", false]], "from_arguments() (tensorrt_llm.models.speculativedecodingmode static method)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.from_arguments", false]], "from_checkpoint() (tensorrt_llm.models.pretrainedconfig class method)": [[84, "tensorrt_llm.models.PretrainedConfig.from_checkpoint", false]], "from_checkpoint() (tensorrt_llm.models.pretrainedmodel class method)": [[84, "tensorrt_llm.models.PretrainedModel.from_checkpoint", false]], "from_config() (tensorrt_llm.models.pretrainedmodel class method)": [[84, "tensorrt_llm.models.PretrainedModel.from_config", false]], "from_dict() (tensorrt_llm.llmapi.buildconfig class method)": [[70, "tensorrt_llm.llmapi.BuildConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.calibconfig class method)": [[70, "tensorrt_llm.llmapi.CalibConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.eagledecodingconfig class method)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.lookaheaddecodingconfig class method)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.medusadecodingconfig class method)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.mtpdecodingconfig class method)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.ngramdecodingconfig class method)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.from_dict", false]], "from_dict() (tensorrt_llm.llmapi.quantconfig class method)": [[70, "tensorrt_llm.llmapi.QuantConfig.from_dict", false]], "from_dict() (tensorrt_llm.models.pretrainedconfig class method)": [[84, "tensorrt_llm.models.PretrainedConfig.from_dict", false]], "from_dir() (tensorrt_llm.runtime.modelrunner class method)": [[87, "tensorrt_llm.runtime.ModelRunner.from_dir", false]], "from_dir() (tensorrt_llm.runtime.modelrunnercpp class method)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.from_dir", false]], "from_engine() (tensorrt_llm.runtime.encdecmodelrunner class method)": [[87, "tensorrt_llm.runtime.EncDecModelRunner.from_engine", false]], "from_engine() (tensorrt_llm.runtime.modelrunner class method)": [[87, "tensorrt_llm.runtime.ModelRunner.from_engine", false]], "from_engine() (tensorrt_llm.runtime.session static method)": [[87, "tensorrt_llm.runtime.Session.from_engine", false]], "from_hugging_face() (tensorrt_llm.models.baichuanforcausallm class method)": [[84, "tensorrt_llm.models.BaichuanForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.chatglmconfig class method)": [[84, "tensorrt_llm.models.ChatGLMConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.chatglmforcausallm class method)": [[84, "tensorrt_llm.models.ChatGLMForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.cogvlmforcausallm class method)": [[84, "tensorrt_llm.models.CogVLMForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.cohereforcausallm class method)": [[84, "tensorrt_llm.models.CohereForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.deepseekforcausallm class method)": [[84, "tensorrt_llm.models.DeepseekForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.deepseekv2forcausallm class method)": [[84, "tensorrt_llm.models.DeepseekV2ForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.eagleforcausallm class method)": [[84, "tensorrt_llm.models.EagleForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.falconconfig class method)": [[84, "tensorrt_llm.models.FalconConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.falconforcausallm class method)": [[84, "tensorrt_llm.models.FalconForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gemmaconfig class method)": [[84, "tensorrt_llm.models.GemmaConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gemmaforcausallm class method)": [[84, "tensorrt_llm.models.GemmaForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptconfig class method)": [[84, "tensorrt_llm.models.GPTConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptforcausallm class method)": [[84, "tensorrt_llm.models.GPTForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptjconfig class method)": [[84, "tensorrt_llm.models.GPTJConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.gptjforcausallm class method)": [[84, "tensorrt_llm.models.GPTJForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llamaconfig class method)": [[84, "tensorrt_llm.models.LLaMAConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llamaforcausallm class method)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llavanextvisionconfig class method)": [[84, "tensorrt_llm.models.LlavaNextVisionConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.llavanextvisionwrapper class method)": [[84, "tensorrt_llm.models.LlavaNextVisionWrapper.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.mambaforcausallm class method)": [[84, "tensorrt_llm.models.MambaForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.medusaconfig class method)": [[84, "tensorrt_llm.models.MedusaConfig.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.medusaforcausallm class method)": [[84, "tensorrt_llm.models.MedusaForCausalLm.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.mllamaforcausallm class method)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.phi3forcausallm class method)": [[84, "tensorrt_llm.models.Phi3ForCausalLM.from_hugging_face", false]], "from_hugging_face() (tensorrt_llm.models.phiforcausallm class method)": [[84, "tensorrt_llm.models.PhiForCausalLM.from_hugging_face", false]], "from_json_file() (tensorrt_llm.llmapi.buildconfig class method)": [[70, "tensorrt_llm.llmapi.BuildConfig.from_json_file", false]], "from_json_file() (tensorrt_llm.models.pretrainedconfig class method)": [[84, "tensorrt_llm.models.PretrainedConfig.from_json_file", false]], "from_meta_ckpt() (tensorrt_llm.models.llamaconfig class method)": [[84, "tensorrt_llm.models.LLaMAConfig.from_meta_ckpt", false]], "from_meta_ckpt() (tensorrt_llm.models.llamaforcausallm class method)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.from_meta_ckpt", false]], "from_nemo() (tensorrt_llm.models.gptconfig class method)": [[84, "tensorrt_llm.models.GPTConfig.from_nemo", false]], "from_nemo() (tensorrt_llm.models.gptforcausallm class method)": [[84, "tensorrt_llm.models.GPTForCausalLM.from_nemo", false]], "from_pretrained() (tensorrt_llm.models.sd3transformer2dmodel class method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.from_pretrained", false]], "from_serialized_engine() (tensorrt_llm.runtime.session static method)": [[87, "tensorrt_llm.runtime.Session.from_serialized_engine", false]], "from_string() (tensorrt_llm.functional.positionembeddingtype static method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.from_string", false]], "from_string() (tensorrt_llm.functional.rotaryscalingtype static method)": [[82, "tensorrt_llm.functional.RotaryScalingType.from_string", false]], "fuse_qkv_projections() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.fuse_qkv_projections", false]], "fusedgatedmlp (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.FusedGatedMLP", false]], "fusedgatedmlp (tensorrt_llm.functional.mlptype attribute)": [[82, "tensorrt_llm.functional.MLPType.FusedGatedMLP", false]], "gatedmlp (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.GatedMLP", false]], "gatedmlp (tensorrt_llm.functional.mlptype attribute)": [[82, "tensorrt_llm.functional.MLPType.GatedMLP", false]], "gather() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gather", false]], "gather_context_logits (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.gather_context_logits", false]], "gather_context_logits (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.gather_context_logits", false]], "gather_generation_logits (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.gather_generation_logits", false]], "gather_generation_logits (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.gather_generation_logits", false]], "gather_last_token_logits() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gather_last_token_logits", false]], "gather_nd() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gather_nd", false]], "gegelu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gegelu", false]], "geglu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.geglu", false]], "gelu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gelu", false]], "gemm_allreduce() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gemm_allreduce", false]], "gemm_allreduce_plugin (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.gemm_allreduce_plugin", false]], "gemm_allreduce_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.gemm_allreduce_plugin", false]], "gemm_swiglu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gemm_swiglu", false]], "gemma2_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[84, "tensorrt_llm.models.GemmaConfig.GEMMA2_ADDED_FIELDS", false]], "gemma2_config() (tensorrt_llm.models.gemmaconfig method)": [[84, "tensorrt_llm.models.GemmaConfig.gemma2_config", false]], "gemma3_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[84, "tensorrt_llm.models.GemmaConfig.GEMMA3_ADDED_FIELDS", false]], "gemma3_config() (tensorrt_llm.models.gemmaconfig method)": [[84, "tensorrt_llm.models.GemmaConfig.gemma3_config", false]], "gemma_added_fields (tensorrt_llm.models.gemmaconfig attribute)": [[84, "tensorrt_llm.models.GemmaConfig.GEMMA_ADDED_FIELDS", false]], "gemmaconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GemmaConfig", false]], "gemmaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GemmaForCausalLM", false]], "generate() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.generate", false]], "generate() (tensorrt_llm.runtime.encdecmodelrunner method)": [[87, "tensorrt_llm.runtime.EncDecModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.modelrunner method)": [[87, "tensorrt_llm.runtime.ModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.modelrunnercpp method)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.generate", false]], "generate() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.generate", false]], "generate() (tensorrt_llm.runtime.qwenforcausallmgenerationsession method)": [[87, "tensorrt_llm.runtime.QWenForCausalLMGenerationSession.generate", false]], "generate_alibi_biases() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.generate_alibi_biases", false]], "generate_alibi_slopes() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.generate_alibi_slopes", false]], "generate_async() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.generate_async", false]], "generate_logn_scaling() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.generate_logn_scaling", false]], "generation_logits (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.generation_logits", false]], "generationsequence (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.GenerationSequence", false]], "generationsession (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.GenerationSession", false]], "get_1d_sincos_pos_embed_from_grid() (in module tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.get_1d_sincos_pos_embed_from_grid", false]], "get_2d_sincos_pos_embed() (in module tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.get_2d_sincos_pos_embed", false]], "get_2d_sincos_pos_embed_from_grid() (in module tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.get_2d_sincos_pos_embed_from_grid", false]], "get_audio_features() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.get_audio_features", false]], "get_batch_idx() (tensorrt_llm.runtime.generationsequence method)": [[87, "tensorrt_llm.runtime.GenerationSequence.get_batch_idx", false]], "get_block_offsets() (tensorrt_llm.runtime.kvcachemanager method)": [[87, "tensorrt_llm.runtime.KVCacheManager.get_block_offsets", false]], "get_comm() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.get_comm", false]], "get_config_group() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.get_config_group", false]], "get_context_phase_params() (tensorrt_llm.llmapi.disaggregatedparams method)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.get_context_phase_params", false]], "get_first_past_key_value() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[83, "tensorrt_llm.layers.attention.KeyValueCacheParams.get_first_past_key_value", false]], "get_hf_config() (tensorrt_llm.models.gemmaconfig static method)": [[84, "tensorrt_llm.models.GemmaConfig.get_hf_config", false]], "get_kv_cache_events() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.get_kv_cache_events", false]], "get_kv_cache_events_async() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.get_kv_cache_events_async", false]], "get_next_medusa_tokens() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.get_next_medusa_tokens", false]], "get_num_heads_kv() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.get_num_heads_kv", false]], "get_parent() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.get_parent", false]], "get_pytorch_backend_config() (tensorrt_llm.llmapi.torchllmargs method)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.get_pytorch_backend_config", false]], "get_request_type() (tensorrt_llm.llmapi.disaggregatedparams method)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.get_request_type", false]], "get_rope_index() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.get_rope_index", false]], "get_seq_idx() (tensorrt_llm.runtime.generationsequence method)": [[87, "tensorrt_llm.runtime.GenerationSequence.get_seq_idx", false]], "get_stats() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.get_stats", false]], "get_stats_async() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.get_stats_async", false]], "get_timestep_embedding() (in module tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.get_timestep_embedding", false]], "get_users() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.get_users", false]], "get_visual_features() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.get_visual_features", false]], "get_weight() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.get_weight", false]], "gpt_attention() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gpt_attention", false]], "gpt_attention_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.gpt_attention_plugin", false]], "gptconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTConfig", false]], "gptforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTForCausalLM", false]], "gptjconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTJConfig", false]], "gptjforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTJForCausalLM", false]], "gptjmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTJModel", false]], "gptmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTModel", false]], "gptneoxforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTNeoXForCausalLM", false]], "gptneoxmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.GPTNeoXModel", false]], "gpu_weights_percent (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.gpu_weights_percent", false]], "grammar (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.grammar", false]], "greedy_sampling (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.greedy_sampling", false]], "group_norm() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.group_norm", false]], "group_size (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.group_size", false]], "groupnorm (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.GroupNorm", false]], "groupnorm (tensorrt_llm.functional.layernormtype attribute)": [[82, "tensorrt_llm.functional.LayerNormType.GroupNorm", false]], "gt() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.gt", false]], "guaranteed_no_evict (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[70, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.GUARANTEED_NO_EVICT", false]], "guided_decoding (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.guided_decoding", false]], "guideddecodingparams (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams", false]], "handle_per_step() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.handle_per_step", false]], "has_affine() (tensorrt_llm.functional.allreduceparams method)": [[82, "tensorrt_llm.functional.AllReduceParams.has_affine", false]], "has_bias() (tensorrt_llm.functional.allreduceparams method)": [[82, "tensorrt_llm.functional.AllReduceParams.has_bias", false]], "has_config_group() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.has_config_group", false]], "has_position_embedding (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.has_position_embedding", false]], "has_position_embedding (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.has_position_embedding", false]], "has_scale() (tensorrt_llm.functional.allreduceparams method)": [[82, "tensorrt_llm.functional.AllReduceParams.has_scale", false]], "has_token_type_embedding (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.has_token_type_embedding", false]], "has_token_type_embedding (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.has_token_type_embedding", false]], "has_zero_point (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.has_zero_point", false]], "head_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.head_size", false]], "head_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.head_size", false]], "hidden_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.hidden_size", false]], "hidden_size (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.hidden_size", false]], "host_cache_size (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.host_cache_size", false]], "identity() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.identity", false]], "ignore_eos (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.ignore_eos", false]], "include_stop_str_in_output (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.include_stop_str_in_output", false]], "index (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.index", false]], "index_select() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.index_select", false]], "infer_shapes() (tensorrt_llm.runtime.session method)": [[87, "tensorrt_llm.runtime.Session.infer_shapes", false]], "inflight (tensorrt_llm.llmapi.batchingtype attribute)": [[70, "tensorrt_llm.llmapi.BatchingType.INFLIGHT", false]], "init_audio_encoder() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.init_audio_encoder", false]], "init_image_encoder() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.init_image_encoder", false]], "init_llm() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.init_llm", false]], "init_processor() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.init_processor", false]], "init_tokenizer() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.init_tokenizer", false]], "input_timing_cache (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.input_timing_cache", false]], "int8 (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.INT8", false]], "int_clip() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.int_clip", false]], "interpolate() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.interpolate", false]], "is_alibi() (tensorrt_llm.functional.positionembeddingtype method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.is_alibi", false]], "is_deferred() (tensorrt_llm.functional.positionembeddingtype method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.is_deferred", false]], "is_dynamic() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.is_dynamic", false]], "is_gated_activation() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.is_gated_activation", false]], "is_gemma_2 (tensorrt_llm.models.gemmaconfig property)": [[84, "tensorrt_llm.models.GemmaConfig.is_gemma_2", false]], "is_gemma_3 (tensorrt_llm.models.gemmaconfig property)": [[84, "tensorrt_llm.models.GemmaConfig.is_gemma_3", false]], "is_keep_all (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.is_keep_all", false]], "is_medusa_mode (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.is_medusa_mode", false]], "is_module_excluded_from_quantization() (tensorrt_llm.llmapi.quantconfig method)": [[70, "tensorrt_llm.llmapi.QuantConfig.is_module_excluded_from_quantization", false]], "is_mrope() (tensorrt_llm.functional.positionembeddingtype method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.is_mrope", false]], "is_public_pool (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.is_public_pool", false]], "is_redrafter_mode (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.is_redrafter_mode", false]], "is_rope() (tensorrt_llm.functional.positionembeddingtype method)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.is_rope", false]], "is_trt_wrapper() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.is_trt_wrapper", false]], "is_use_oldest (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.is_use_oldest", false]], "is_valid() (tensorrt_llm.layers.attention.attentionparams method)": [[83, "tensorrt_llm.layers.attention.AttentionParams.is_valid", false]], "is_valid() (tensorrt_llm.layers.attention.keyvaluecacheparams method)": [[83, "tensorrt_llm.layers.attention.KeyValueCacheParams.is_valid", false]], "is_valid_cross_attn() (tensorrt_llm.layers.attention.attentionparams method)": [[83, "tensorrt_llm.layers.attention.AttentionParams.is_valid_cross_attn", false]], "joint_attn_forward() (tensorrt_llm.layers.attention.diffusersattention method)": [[83, "tensorrt_llm.layers.attention.DiffusersAttention.joint_attn_forward", false]], "json (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.json", false]], "json_object (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.json_object", false]], "keyvaluecacheparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.KeyValueCacheParams", false]], "kv_cache_dtype (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.kv_cache_dtype", false]], "kv_cache_quant_algo (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.kv_cache_quant_algo", false]], "kv_cache_type (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.kv_cache_type", false]], "kv_cache_type (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.kv_cache_type", false]], "kv_cache_type (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.kv_cache_type", false]], "kv_dtype (tensorrt_llm.models.pretrainedconfig property)": [[84, "tensorrt_llm.models.PretrainedConfig.kv_dtype", false]], "kvcacheconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.KvCacheConfig", false]], "kvcachemanager (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.KVCacheManager", false]], "kvcacheretentionconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig", false]], "kvcacheretentionconfig.tokenrangeretentionconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig", false]], "labelembedding (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.LabelEmbedding", false]], "language_adapter_config (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.language_adapter_config", false]], "last_layer (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.last_layer", false]], "last_process_for_ub (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.LAST_PROCESS_FOR_UB", false]], "layer_norm() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.layer_norm", false]], "layer_quant_mode (tensorrt_llm.llmapi.quantconfig property)": [[70, "tensorrt_llm.llmapi.QuantConfig.layer_quant_mode", false]], "layer_types (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.layer_types", false]], "layernorm (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.LayerNorm", false]], "layernorm (tensorrt_llm.functional.layernormtype attribute)": [[82, "tensorrt_llm.functional.LayerNormType.LayerNorm", false]], "layernormpositiontype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.LayerNormPositionType", false]], "layernormtype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.LayerNormType", false]], "learned_absolute (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.learned_absolute", false]], "length (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.length", false]], "length (tensorrt_llm.llmapi.completionoutput property)": [[70, "id2", false]], "length_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.length_penalty", false]], "length_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.length_penalty", false]], "linear (class in tensorrt_llm.layers.linear)": [[83, "tensorrt_llm.layers.linear.Linear", false]], "linear (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.linear", false]], "linearactivation (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.LinearActivation", false]], "linearapproximategelu (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.LinearApproximateGELU", false]], "linearbase (class in tensorrt_llm.layers.linear)": [[83, "tensorrt_llm.layers.linear.LinearBase", false]], "lineargeglu (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.LinearGEGLU", false]], "lineargelu (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.LinearGELU", false]], "linearswiglu (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.LinearSwiGLU", false]], "llama3 (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.llama3", false]], "llamaconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.LLaMAConfig", false]], "llamaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.LLaMAForCausalLM", false]], "llamamodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.LLaMAModel", false]], "llavanextvisionconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.LlavaNextVisionConfig", false]], "llavanextvisionwrapper (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.LlavaNextVisionWrapper", false]], "llm (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.LLM", false]], "llm_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.llm_engine_dir", false]], "llmargs (in module tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.LlmArgs", false]], "load() (tensorrt_llm.models.pretrainedmodel method)": [[84, "tensorrt_llm.models.PretrainedModel.load", false]], "load() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.load", false]], "load_format (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.load_format", false]], "load_test_audio() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.load_test_audio", false]], "load_test_data() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.load_test_data", false]], "locate_accepted_draft_tokens() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.locate_accepted_draft_tokens", false]], "location (tensorrt_llm.functional.tensor property)": [[82, "tensorrt_llm.functional.Tensor.location", false]], "log() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.log", false]], "log() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.log", false]], "log_softmax() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.log_softmax", false]], "logits_processor (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.logits_processor", false]], "logitsprocessor (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.LogitsProcessor", false]], "logitsprocessorlist (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.LogitsProcessorList", false]], "logprobs (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.logprobs", false]], "logprobs (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.logprobs", false]], "logprobs_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.logprobs_diff", false]], "logprobs_diff (tensorrt_llm.llmapi.completionoutput property)": [[70, "id3", false]], "long_rope (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.long_rope", false]], "longrope (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.longrope", false]], "lookahead_config (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.lookahead_config", false]], "lookahead_decoding (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.LOOKAHEAD_DECODING", false]], "lookaheaddecodingconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig", false]], "lora_config (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.lora_config", false]], "lora_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.lora_plugin", false]], "lora_plugin() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.lora_plugin", false]], "lora_target_modules (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.lora_target_modules", false]], "low_latency_gemm() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.low_latency_gemm", false]], "low_latency_gemm_swiglu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.low_latency_gemm_swiglu", false]], "lowprecision (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.LOWPRECISION", false]], "lt() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.lt", false]], "make_causal_mask() (in module tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.make_causal_mask", false]], "mamba_conv1d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.mamba_conv1d", false]], "mamba_conv1d_plugin (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.mamba_conv1d_plugin", false]], "mambaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MambaForCausalLM", false]], "mapping (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.mapping", false]], "mapping (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.mapping", false]], "mark_output() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.mark_output", false]], "masked_scatter() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.masked_scatter", false]], "masked_select() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.masked_select", false]], "matmul() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.matmul", false]], "max() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.max", false]], "max() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.max", false]], "max_attention_window (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.max_attention_window", false]], "max_attention_window_size (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.max_attention_window_size", false]], "max_batch_size (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_batch_size", false]], "max_batch_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.max_batch_size", false]], "max_beam_width (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_beam_width", false]], "max_beam_width (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.max_beam_width", false]], "max_cache_storage_gb (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildCacheConfig.max_cache_storage_gb", false]], "max_cache_storage_gb (tensorrt_llm.llmapi.buildcacheconfig property)": [[70, "id8", false]], "max_cpu_loras (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.max_cpu_loras", false]], "max_cpu_loras (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.max_cpu_loras", false]], "max_draft_len (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_draft_len", false]], "max_draft_tokens (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.max_draft_tokens", false]], "max_encoder_input_len (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_encoder_input_len", false]], "max_input_len (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_input_len", false]], "max_lora_rank (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.max_lora_rank", false]], "max_lora_rank (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.max_lora_rank", false]], "max_loras (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.max_loras", false]], "max_loras (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.max_loras", false]], "max_matching_ngram_size (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.max_matching_ngram_size", false]], "max_medusa_tokens (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.max_medusa_tokens", false]], "max_new_tokens (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.max_new_tokens", false]], "max_ngram_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_ngram_size", false]], "max_non_leaves_per_layer (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.max_non_leaves_per_layer", false]], "max_num_tokens (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_num_tokens", false]], "max_num_tokens (tensorrt_llm.llmapi.cachetransceiverconfig attribute)": [[70, "tensorrt_llm.llmapi.CacheTransceiverConfig.max_num_tokens", false]], "max_prompt_embedding_table_size (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.max_prompt_embedding_table_size", false]], "max_prompt_embedding_table_size (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.max_prompt_embedding_table_size", false]], "max_records (tensorrt_llm.llmapi.buildcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildCacheConfig.max_records", false]], "max_records (tensorrt_llm.llmapi.buildcacheconfig property)": [[70, "id9", false]], "max_seq_len (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.max_seq_len", false]], "max_sequence_length (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.max_sequence_length", false]], "max_sequence_length (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.max_sequence_length", false]], "max_tokens (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.max_tokens", false]], "max_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.max_tokens", false]], "max_utilization (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[70, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.MAX_UTILIZATION", false]], "max_verification_set_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_verification_set_size", false]], "max_window_size (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.max_window_size", false]], "maximum() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.maximum", false]], "mean() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.mean", false]], "mean() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.mean", false]], "medusa (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.MEDUSA", false]], "medusa_choices (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig.medusa_choices", false]], "medusa_decode_and_verify() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_decode_and_verify", false]], "medusa_paths (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_paths", false]], "medusa_position_offsets (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_position_offsets", false]], "medusa_temperature (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_temperature", false]], "medusa_topks (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_topks", false]], "medusa_tree_ids (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.medusa_tree_ids", false]], "medusaconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MedusaConfig", false]], "medusadecodingconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig", false]], "medusaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MedusaForCausalLm", false]], "meshgrid2d() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.meshgrid2d", false]], "min() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.min", false]], "min_latency (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.MIN_LATENCY", false]], "min_length (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.min_length", false]], "min_p (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.min_p", false]], "min_p (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.min_p", false]], "min_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.min_tokens", false]], "minimum() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.minimum", false]], "mish (class in tensorrt_llm.layers.activation)": [[83, "tensorrt_llm.layers.activation.Mish", false]], "mixed_precision (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.MIXED_PRECISION", false]], "mixed_sampler (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.mixed_sampler", false]], "mllamaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM", false]], "mlp (class in tensorrt_llm.layers.mlp)": [[83, "tensorrt_llm.layers.mlp.MLP", false]], "mlp (tensorrt_llm.functional.mlptype attribute)": [[82, "tensorrt_llm.functional.MLPType.MLP", false]], "mlptype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.MLPType", false]], "model": [[30, "cmdoption-trtllm-serve-serve-arg-MODEL", false]], "model_config (tensorrt_llm.llmapi.cachetransceiverconfig attribute)": [[70, "tensorrt_llm.llmapi.CacheTransceiverConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.dynamicbatchconfig attribute)": [[70, "tensorrt_llm.llmapi.DynamicBatchConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.lookaheaddecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.schedulerconfig attribute)": [[70, "tensorrt_llm.llmapi.SchedulerConfig.model_config", false]], "model_config (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.model_config", false]], "model_config (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.model_config", false]], "model_name (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.model_name", false]], "model_post_init() (tensorrt_llm.llmapi.torchllmargs method)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.model_post_init", false]], "model_post_init() (tensorrt_llm.llmapi.trtllmargs method)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.model_post_init", false]], "modelconfig (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.ModelConfig", false]], "modelrunner (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.ModelRunner", false]], "modelrunnercpp (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp", false]], "module": [[82, "module-tensorrt_llm", false], [82, "module-tensorrt_llm.functional", false], [83, "module-tensorrt_llm", false], [83, "module-tensorrt_llm.layers.activation", false], [83, "module-tensorrt_llm.layers.attention", false], [83, "module-tensorrt_llm.layers.cast", false], [83, "module-tensorrt_llm.layers.conv", false], [83, "module-tensorrt_llm.layers.embedding", false], [83, "module-tensorrt_llm.layers.linear", false], [83, "module-tensorrt_llm.layers.mlp", false], [83, "module-tensorrt_llm.layers.normalization", false], [83, "module-tensorrt_llm.layers.pooling", false], [84, "module-tensorrt_llm", false], [84, "module-tensorrt_llm.models", false], [85, "module-tensorrt_llm", false], [85, "module-tensorrt_llm.plugin", false], [86, "module-tensorrt_llm", false], [86, "module-tensorrt_llm.quantization", false], [87, "module-tensorrt_llm", false], [87, "module-tensorrt_llm.runtime", false]], "modulo() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.modulo", false]], "moe (tensorrt_llm.functional.sidestreamidtype attribute)": [[82, "tensorrt_llm.functional.SideStreamIDType.moe", false]], "moe_allreduce_residual_rms_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.MOE_ALLREDUCE_RESIDUAL_RMS_NORM", false]], "moe_backend (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.moe_backend", false]], "moe_load_balancer (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.moe_load_balancer", false]], "moe_max_num_tokens (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.moe_max_num_tokens", false]], "monitor_memory (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.monitor_memory", false]], "mpicommsession (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.MpiCommSession", false]], "mptforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MPTForCausalLM", false]], "mptmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.MPTModel", false]], "mrope (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.mrope", false]], "mrope (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.mrope", false]], "mropeparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.MropeParams", false]], "msg (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "id10", false], [70, "id13", false], [70, "id16", false], [70, "tensorrt_llm.llmapi.TorchLlmArgs.msg", false]], "msg (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "id19", false], [70, "id22", false], [70, "id25", false], [70, "id28", false], [70, "id31", false], [70, "tensorrt_llm.llmapi.TrtLlmArgs.msg", false]], "mtpdecodingconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig", false]], "mul() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.mul", false]], "multi_block_mode (tensorrt_llm.llmapi.extendedruntimeperfknobconfig attribute)": [[70, "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig.multi_block_mode", false]], "multimodalmodelrunner (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner", false]], "multiply_and_lora() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.multiply_and_lora", false]], "multiply_collect() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.multiply_collect", false]], "multiply_collect() (tensorrt_llm.layers.linear.rowlinear method)": [[83, "tensorrt_llm.layers.linear.RowLinear.multiply_collect", false]], "n (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.n", false]], "name (tensorrt_llm.functional.tensor property)": [[82, "tensorrt_llm.functional.Tensor.name", false]], "name (tensorrt_llm.runtime.tensorinfo attribute)": [[87, "tensorrt_llm.runtime.TensorInfo.name", false]], "native_quant_flow (tensorrt_llm.models.gemmaforcausallm attribute)": [[84, "tensorrt_llm.models.GemmaForCausalLM.NATIVE_QUANT_FLOW", false]], "nccl (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.NCCL", false]], "ndim() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.ndim", false]], "network (tensorrt_llm.functional.tensor property)": [[82, "tensorrt_llm.functional.Tensor.network", false]], "next_medusa_input_ids() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.next_medusa_input_ids", false]], "ngram (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.NGRAM", false]], "ngramdecodingconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig", false]], "no_quant (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.NO_QUANT", false]], "no_repeat_ngram_size (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.no_repeat_ngram_size", false]], "no_repeat_ngram_size (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.no_repeat_ngram_size", false]], "non_gated_version() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.non_gated_version", false]], "none (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.NONE", false]], "none (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.none", false]], "none (tensorrt_llm.models.speculativedecodingmode attribute)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode.NONE", false]], "nonzero() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.nonzero", false]], "not_op() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.not_op", false]], "num_beams (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.num_beams", false]], "num_draft_tokens (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.num_draft_tokens", false]], "num_eagle_layers (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.num_eagle_layers", false]], "num_heads (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.num_heads", false]], "num_heads (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.num_heads", false]], "num_kv_heads (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_kv_heads", false]], "num_kv_heads_per_cross_attn_layer (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_kv_heads_per_cross_attn_layer", false]], "num_kv_heads_per_layer (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_kv_heads_per_layer", false]], "num_layers (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.num_layers", false]], "num_layers (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.num_layers", false]], "num_medusa_heads (tensorrt_llm.llmapi.medusadecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MedusaDecodingConfig.num_medusa_heads", false]], "num_medusa_heads (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.num_medusa_heads", false]], "num_medusa_heads (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.num_medusa_heads", false]], "num_nextn_predict_layers (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.num_nextn_predict_layers", false]], "num_return_sequences (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.num_return_sequences", false]], "numel() (tensorrt_llm.runtime.tensorinfo method)": [[87, "tensorrt_llm.runtime.TensorInfo.numel", false]], "nvfp4 (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.NVFP4", false]], "nvinfer1 (c++ type)": [[1, "_CPPv48nvinfer1", false]], "onboard_blocks (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.onboard_blocks", false]], "oneshot (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.ONESHOT", false]], "op_and() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.op_and", false]], "op_or() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.op_or", false]], "op_xor() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.op_xor", false]], "opaque_state (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.opaque_state", false]], "opt_batch_size (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.opt_batch_size", false]], "opt_num_tokens (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.opt_num_tokens", false]], "optforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.OPTForCausalLM", false]], "optmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.OPTModel", false]], "outer() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.outer", false]], "output_cum_log_probs (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.output_cum_log_probs", false]], "output_log_probs (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.output_log_probs", false]], "output_sequence_lengths (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.output_sequence_lengths", false]], "output_timing_cache (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.output_timing_cache", false]], "outputs (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.outputs", false]], "pad() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.pad", false]], "pad_id (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.pad_id", false]], "pad_id (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.pad_id", false]], "padding (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.padding", false]], "paged_kv_cache (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.paged_kv_cache", false]], "paged_state (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.paged_state", false]], "paged_state (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.paged_state", false]], "permute() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.permute", false]], "permute() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.permute", false]], "phi3forcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.Phi3ForCausalLM", false]], "phi3model (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.Phi3Model", false]], "phiforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.PhiForCausalLM", false]], "phimodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.PhiModel", false]], "pixartalphatextprojection (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection", false]], "plugin_config (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.plugin_config", false]], "pluginconfig (class in tensorrt_llm.plugin)": [[85, "tensorrt_llm.plugin.PluginConfig", false]], "positionembeddingtype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.PositionEmbeddingType", false]], "post_layernorm (tensorrt_llm.functional.layernormpositiontype attribute)": [[82, "tensorrt_llm.functional.LayerNormPositionType.post_layernorm", false]], "posterior_threshold (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.posterior_threshold", false]], "postprocess() (tensorrt_llm.layers.attention.attention method)": [[83, "tensorrt_llm.layers.attention.Attention.postprocess", false]], "postprocess() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[83, "tensorrt_llm.layers.attention.DeepseekV2Attention.postprocess", false]], "postprocess() (tensorrt_llm.layers.embedding.embedding method)": [[83, "tensorrt_llm.layers.embedding.Embedding.postprocess", false]], "postprocess() (tensorrt_llm.layers.linear.linear method)": [[83, "tensorrt_llm.layers.linear.Linear.postprocess", false]], "pow() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.pow", false]], "pp_communicate_final_output_ids() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.pp_communicate_final_output_ids", false]], "pp_communicate_new_tokens() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.pp_communicate_new_tokens", false]], "pre_layernorm (tensorrt_llm.functional.layernormpositiontype attribute)": [[82, "tensorrt_llm.functional.LayerNormPositionType.pre_layernorm", false]], "pre_quant_scale (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.pre_quant_scale", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.decodermodel method)": [[84, "tensorrt_llm.models.DecoderModel.precompute_relative_attention_bias", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.precompute_relative_attention_bias", false]], "precompute_relative_attention_bias() (tensorrt_llm.models.whisperencoder method)": [[84, "tensorrt_llm.models.WhisperEncoder.precompute_relative_attention_bias", false]], "prepare_inputs() (tensorrt_llm.models.chatglmforcausallm method)": [[84, "tensorrt_llm.models.ChatGLMForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.decodermodel method)": [[84, "tensorrt_llm.models.DecoderModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.eagleforcausallm method)": [[84, "tensorrt_llm.models.EagleForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.llavanextvisionwrapper method)": [[84, "tensorrt_llm.models.LlavaNextVisionWrapper.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.mambaforcausallm method)": [[84, "tensorrt_llm.models.MambaForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.mllamaforcausallm method)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.pretrainedmodel method)": [[84, "tensorrt_llm.models.PretrainedModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[84, "tensorrt_llm.models.RecurrentGemmaForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.redrafterforcausallm method)": [[84, "tensorrt_llm.models.ReDrafterForCausalLM.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.prepare_inputs", false]], "prepare_inputs() (tensorrt_llm.models.whisperencoder method)": [[84, "tensorrt_llm.models.WhisperEncoder.prepare_inputs", false]], "prepare_position_ids_for_cogvlm() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.prepare_position_ids_for_cogvlm", false]], "prepare_recurrent_inputs() (tensorrt_llm.models.recurrentgemmaforcausallm method)": [[84, "tensorrt_llm.models.RecurrentGemmaForCausalLM.prepare_recurrent_inputs", false]], "preprocess() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.preprocess", false]], "presence_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.presence_penalty", false]], "presence_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.presence_penalty", false]], "pretrainedconfig (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.PretrainedConfig", false]], "pretrainedmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.PretrainedModel", false]], "print_iter_log (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.print_iter_log", false]], "priority (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.priority", false]], "process_input() (tensorrt_llm.runtime.encdecmodelrunner method)": [[87, "tensorrt_llm.runtime.EncDecModelRunner.process_input", false]], "process_logits_including_draft() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.process_logits_including_draft", false]], "prod() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.prod", false]], "profiler (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.profiler", false]], "profiling_verbosity (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.profiling_verbosity", false]], "prompt (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.prompt", false]], "prompt (tensorrt_llm.llmapi.requestoutput property)": [[70, "id6", false]], "prompt_logprobs (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.prompt_logprobs", false]], "prompt_logprobs (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.prompt_logprobs", false]], "prompt_lookup_num_tokens (tensorrt_llm.llmapi.ngramdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.NGramDecodingConfig.prompt_lookup_num_tokens", false]], "prompt_token_ids (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.prompt_token_ids", false]], "prompttuningembedding (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.PromptTuningEmbedding", false]], "ptuning_setup() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup", false]], "ptuning_setup_fuyu() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_fuyu", false]], "ptuning_setup_llava_next() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_llava_next", false]], "ptuning_setup_phi3() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_phi3", false]], "ptuning_setup_pixtral() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.ptuning_setup_pixtral", false]], "python_e2e (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.python_e2e", false]], "pytorch_eagle_weights_path (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.pytorch_eagle_weights_path", false]], "quant_algo (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.quant_algo", false]], "quant_algo (tensorrt_llm.models.pretrainedconfig property)": [[84, "tensorrt_llm.models.PretrainedConfig.quant_algo", false]], "quant_mode (tensorrt_llm.llmapi.quantconfig property)": [[70, "tensorrt_llm.llmapi.QuantConfig.quant_mode", false]], "quant_mode (tensorrt_llm.models.pretrainedconfig property)": [[84, "tensorrt_llm.models.PretrainedConfig.quant_mode", false]], "quant_mode (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.quant_mode", false]], "quant_mode (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.quant_mode", false]], "quantalgo (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.QuantAlgo", false]], "quantalgo (class in tensorrt_llm.quantization)": [[86, "tensorrt_llm.quantization.QuantAlgo", false]], "quantconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.QuantConfig", false]], "quantize() (tensorrt_llm.models.baichuanforcausallm class method)": [[84, "tensorrt_llm.models.BaichuanForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.chatglmforcausallm class method)": [[84, "tensorrt_llm.models.ChatGLMForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.cogvlmforcausallm class method)": [[84, "tensorrt_llm.models.CogVLMForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.gemmaforcausallm class method)": [[84, "tensorrt_llm.models.GemmaForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.gptforcausallm class method)": [[84, "tensorrt_llm.models.GPTForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.llamaforcausallm class method)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.quantize", false]], "quantize() (tensorrt_llm.models.pretrainedmodel class method)": [[84, "tensorrt_llm.models.PretrainedModel.quantize", false]], "quantize_and_export() (in module tensorrt_llm.quantization)": [[86, "tensorrt_llm.quantization.quantize_and_export", false]], "quantmode (class in tensorrt_llm.quantization)": [[86, "tensorrt_llm.quantization.QuantMode", false]], "quick_gelu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.quick_gelu", false]], "qwenforcausallmgenerationsession (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.QWenForCausalLMGenerationSession", false]], "rand() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.rand", false]], "random_seed (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.random_seed", false]], "random_seed (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.random_seed", false]], "rank() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.rank", false]], "rearrange() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.rearrange", false]], "recurrentgemmaforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.RecurrentGemmaForCausalLM", false]], "recv() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.recv", false]], "redrafter_draft_len_per_beam (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.redrafter_draft_len_per_beam", false]], "redrafter_num_beams (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.redrafter_num_beams", false]], "redrafterforcausallm (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.ReDrafterForCausalLM", false]], "reduce() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.reduce", false]], "reduce_scatter() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.reduce_scatter", false]], "regex (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.regex", false]], "relative (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.relative", false]], "relaxed_delta (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.relaxed_delta", false]], "relaxed_topk (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.relaxed_topk", false]], "release() (tensorrt_llm.models.pretrainedmodel method)": [[84, "tensorrt_llm.models.PretrainedModel.release", false]], "relu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.relu", false]], "remove_input_padding (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.remove_input_padding", false]], "remove_input_padding (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.remove_input_padding", false]], "reorder_kv_cache_for_beam_search() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.reorder_kv_cache_for_beam_search", false]], "repeat() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.repeat", false]], "repeat() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.repeat", false]], "repeat_interleave() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.repeat_interleave", false]], "repetition_penalty (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.repetition_penalty", false]], "repetition_penalty (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.repetition_penalty", false]], "replace_all_uses_with() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.replace_all_uses_with", false]], "request_id (tensorrt_llm.llmapi.requestoutput attribute)": [[70, "tensorrt_llm.llmapi.RequestOutput.request_id", false]], "request_type (tensorrt_llm.llmapi.disaggregatedparams attribute)": [[70, "tensorrt_llm.llmapi.DisaggregatedParams.request_type", false]], "requesterror (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.RequestError", false]], "requestoutput (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.RequestOutput", false]], "residual_rms_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM", false]], "residual_rms_norm_out_quant_fp8 (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_OUT_QUANT_FP8", false]], "residual_rms_norm_out_quant_nvfp4 (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_OUT_QUANT_NVFP4", false]], "residual_rms_norm_quant_fp8 (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_QUANT_FP8", false]], "residual_rms_norm_quant_nvfp4 (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_NORM_QUANT_NVFP4", false]], "residual_rms_prepost_norm (tensorrt_llm.functional.allreducefusionop attribute)": [[82, "tensorrt_llm.functional.AllReduceFusionOp.RESIDUAL_RMS_PREPOST_NORM", false]], "return_context_logits (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.return_context_logits", false]], "return_dict (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.return_dict", false]], "return_encoder_output (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.return_encoder_output", false]], "return_generation_logits (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.return_generation_logits", false]], "return_perf_metrics (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.return_perf_metrics", false]], "rg_lru() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.rg_lru", false]], "rms_norm() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.rms_norm", false]], "rmsnorm (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.RmsNorm", false]], "rmsnorm (tensorrt_llm.functional.layernormtype attribute)": [[82, "tensorrt_llm.functional.LayerNormType.RmsNorm", false]], "rnn_conv_dim_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.rnn_conv_dim_size", false]], "rnn_conv_dim_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.rnn_conv_dim_size", false]], "rnn_head_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.rnn_head_size", false]], "rnn_head_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.rnn_head_size", false]], "rnn_hidden_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.rnn_hidden_size", false]], "rnn_hidden_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.rnn_hidden_size", false]], "robertaforquestionanswering (in module tensorrt_llm.models)": [[84, "tensorrt_llm.models.RobertaForQuestionAnswering", false]], "robertaforsequenceclassification (in module tensorrt_llm.models)": [[84, "tensorrt_llm.models.RobertaForSequenceClassification", false]], "robertamodel (in module tensorrt_llm.models)": [[84, "tensorrt_llm.models.RobertaModel", false]], "rope_gpt_neox (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.rope_gpt_neox", false]], "rope_gptj (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.rope_gptj", false]], "ropeembeddingutils (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils", false]], "rotaryscalingtype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.RotaryScalingType", false]], "rotate_every_two() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.rotate_every_two", false]], "rotate_half() (tensorrt_llm.functional.ropeembeddingutils static method)": [[82, "tensorrt_llm.functional.RopeEmbeddingUtils.rotate_half", false]], "round() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.round", false]], "rowlinear (class in tensorrt_llm.layers.linear)": [[83, "tensorrt_llm.layers.linear.RowLinear", false]], "run() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.run", false]], "run() (tensorrt_llm.runtime.session method)": [[87, "tensorrt_llm.runtime.Session.run", false]], "runtime (tensorrt_llm.runtime.generationsession attribute)": [[87, "tensorrt_llm.runtime.GenerationSession.runtime", false]], "runtime (tensorrt_llm.runtime.session property)": [[87, "tensorrt_llm.runtime.Session.runtime", false]], "samplingconfig (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.SamplingConfig", false]], "samplingparams (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.SamplingParams", false]], "save() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.save", false]], "save_checkpoint() (tensorrt_llm.models.llavanextvisionwrapper method)": [[84, "tensorrt_llm.models.LlavaNextVisionWrapper.save_checkpoint", false]], "save_checkpoint() (tensorrt_llm.models.pretrainedmodel method)": [[84, "tensorrt_llm.models.PretrainedModel.save_checkpoint", false]], "scatter() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.scatter", false]], "scatter_nd() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.scatter_nd", false]], "schedulerconfig (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.SchedulerConfig", false]], "sd35adalayernormzerox (class in tensorrt_llm.layers.normalization)": [[83, "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX", false]], "sd3patchembed (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.SD3PatchEmbed", false]], "sd3transformer2dmodel (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.SD3Transformer2DModel", false]], "secondary_offload_min_priority (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.secondary_offload_min_priority", false]], "seed (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.seed", false]], "select() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.select", false]], "select() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.select", false]], "selective_scan() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.selective_scan", false]], "send() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.send", false]], "serialize_engine() (tensorrt_llm.runtime.modelrunner method)": [[87, "tensorrt_llm.runtime.ModelRunner.serialize_engine", false]], "session (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.Session", false]], "set_attn_processor() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.set_attn_processor", false]], "set_from_optional (c macro)": [[1, "c.SET_FROM_OPTIONAL", false]], "set_if_not_exist() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.set_if_not_exist", false]], "set_rank() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.set_rank", false]], "set_rel_attn_table() (tensorrt_llm.layers.attention.attention method)": [[83, "tensorrt_llm.layers.attention.Attention.set_rel_attn_table", false]], "set_shapes() (tensorrt_llm.runtime.session method)": [[87, "tensorrt_llm.runtime.Session.set_shapes", false]], "setup() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.setup", false]], "setup_fake_prompts() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts", false]], "setup_fake_prompts_qwen2vl() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts_qwen2vl", false]], "setup_fake_prompts_vila() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.setup_fake_prompts_vila", false]], "setup_inputs() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.setup_inputs", false]], "shape (tensorrt_llm.functional.tensor property)": [[82, "tensorrt_llm.functional.Tensor.shape", false]], "shape (tensorrt_llm.runtime.tensorinfo attribute)": [[87, "tensorrt_llm.runtime.TensorInfo.shape", false]], "shape() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.shape", false]], "shutdown() (tensorrt_llm.llmapi.llm method)": [[70, "tensorrt_llm.llmapi.LLM.shutdown", false]], "shutdown() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.shutdown", false]], "sidestreamidtype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.SideStreamIDType", false]], "sigmoid() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.sigmoid", false]], "silu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.silu", false]], "sin() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.sin", false]], "sink_token_length (tensorrt_llm.llmapi.kvcacheconfig attribute)": [[70, "tensorrt_llm.llmapi.KvCacheConfig.sink_token_length", false]], "sink_token_length (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.sink_token_length", false]], "size (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.size", false]], "size() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.size", false]], "skip_cross_attn_blocks (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.skip_cross_attn_blocks", false]], "skip_cross_kv (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.skip_cross_kv", false]], "skip_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.skip_special_tokens", false]], "slice() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.slice", false]], "sliceinputtype (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.SliceInputType", false]], "sliding_window_causal (tensorrt_llm.functional.attentionmasktype attribute)": [[82, "tensorrt_llm.functional.AttentionMaskType.sliding_window_causal", false]], "smoothquant_val (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.smoothquant_val", false]], "softmax() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.softmax", false]], "softplus() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.softplus", false]], "spaces_between_special_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.spaces_between_special_tokens", false]], "specdecodingparams (class in tensorrt_llm.layers.attention)": [[83, "tensorrt_llm.layers.attention.SpecDecodingParams", false]], "speculative_decoding_mode (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.speculative_decoding_mode", false]], "speculativedecodingmode (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.SpeculativeDecodingMode", false]], "split() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.split", false]], "split() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.split", false]], "split_prompt_by_images() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.split_prompt_by_images", false]], "sqrt() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.sqrt", false]], "sqrt() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.sqrt", false]], "squared_relu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.squared_relu", false]], "squeeze() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.squeeze", false]], "squeeze() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.squeeze", false]], "squeeze() (tensorrt_llm.runtime.tensorinfo method)": [[87, "tensorrt_llm.runtime.TensorInfo.squeeze", false]], "stack() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.stack", false]], "start (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.start", false]], "state_dtype (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.state_dtype", false]], "state_dtype (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.state_dtype", false]], "state_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.state_size", false]], "state_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.state_size", false]], "static (tensorrt_llm.llmapi.batchingtype attribute)": [[70, "tensorrt_llm.llmapi.BatchingType.STATIC", false]], "static_batch (tensorrt_llm.llmapi.capacityschedulerpolicy attribute)": [[70, "tensorrt_llm.llmapi.CapacitySchedulerPolicy.STATIC_BATCH", false]], "step() (tensorrt_llm.runtime.kvcachemanager method)": [[87, "tensorrt_llm.runtime.KVCacheManager.step", false]], "stop (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.stop", false]], "stop_reason (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.stop_reason", false]], "stop_token_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.stop_token_ids", false]], "stop_words_list (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.stop_words_list", false]], "stoppingcriteria (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.StoppingCriteria", false]], "stoppingcriterialist (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.StoppingCriteriaList", false]], "stride (tensorrt_llm.functional.sliceinputtype attribute)": [[82, "tensorrt_llm.functional.SliceInputType.stride", false]], "strongly_typed (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.strongly_typed", false]], "structural_tag (tensorrt_llm.llmapi.guideddecodingparams attribute)": [[70, "tensorrt_llm.llmapi.GuidedDecodingParams.structural_tag", false]], "sub() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.sub", false]], "submit() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.submit", false]], "submit_sync() (tensorrt_llm.llmapi.mpicommsession method)": [[70, "tensorrt_llm.llmapi.MpiCommSession.submit_sync", false]], "sum() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.sum", false]], "swiglu() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.swiglu", false]], "tanh() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.tanh", false]], "temperature (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.temperature", false]], "temperature (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.temperature", false]], "tensor (class in tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.Tensor", false]], "tensorinfo (class in tensorrt_llm.runtime)": [[87, "tensorrt_llm.runtime.TensorInfo", false]], "tensorrt_llm": [[82, "module-tensorrt_llm", false], [83, "module-tensorrt_llm", false], [84, "module-tensorrt_llm", false], [85, "module-tensorrt_llm", false], [86, "module-tensorrt_llm", false], [87, "module-tensorrt_llm", false]], "tensorrt_llm (c++ type)": [[0, "_CPPv412tensorrt_llm", false], [1, "_CPPv412tensorrt_llm", false]], "tensorrt_llm.functional": [[82, "module-tensorrt_llm.functional", false]], "tensorrt_llm.layers.activation": [[83, "module-tensorrt_llm.layers.activation", false]], "tensorrt_llm.layers.attention": [[83, "module-tensorrt_llm.layers.attention", false]], "tensorrt_llm.layers.cast": [[83, "module-tensorrt_llm.layers.cast", false]], "tensorrt_llm.layers.conv": [[83, "module-tensorrt_llm.layers.conv", false]], "tensorrt_llm.layers.embedding": [[83, "module-tensorrt_llm.layers.embedding", false]], "tensorrt_llm.layers.linear": [[83, "module-tensorrt_llm.layers.linear", false]], "tensorrt_llm.layers.mlp": [[83, "module-tensorrt_llm.layers.mlp", false]], "tensorrt_llm.layers.normalization": [[83, "module-tensorrt_llm.layers.normalization", false]], "tensorrt_llm.layers.pooling": [[83, "module-tensorrt_llm.layers.pooling", false]], "tensorrt_llm.models": [[84, "module-tensorrt_llm.models", false]], "tensorrt_llm.plugin": [[85, "module-tensorrt_llm.plugin", false]], "tensorrt_llm.quantization": [[86, "module-tensorrt_llm.quantization", false]], "tensorrt_llm.runtime": [[87, "module-tensorrt_llm.runtime", false]], "tensorrt_llm::batch_manager (c++ type)": [[0, "_CPPv4N12tensorrt_llm13batch_managerE", false], [1, "_CPPv4N12tensorrt_llm13batch_managerE", false]], "tensorrt_llm::batch_manager::kv_cache_manager (c++ type)": [[0, "_CPPv4N12tensorrt_llm13batch_manager16kv_cache_managerE", false]], "tensorrt_llm::executor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executorE", false]], "tensorrt_llm::executor::additionalmodeloutput (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutputE", false]], "tensorrt_llm::executor::additionalmodeloutput::additionalmodeloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", false]], "tensorrt_llm::executor::additionalmodeloutput::gathercontext (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput13gatherContextE", false]], "tensorrt_llm::executor::additionalmodeloutput::name (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput4nameE", false]], "tensorrt_llm::executor::additionalmodeloutput::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", false]], "tensorrt_llm::executor::additionaloutput (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputE", false]], "tensorrt_llm::executor::additionaloutput::additionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", false]], "tensorrt_llm::executor::additionaloutput::name (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput4nameE", false]], "tensorrt_llm::executor::additionaloutput::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", false]], "tensorrt_llm::executor::additionaloutput::output (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput6outputE", false]], "tensorrt_llm::executor::additionaloutput::~additionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputD0Ev", false]], "tensorrt_llm::executor::batchingtype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingTypeE", false]], "tensorrt_llm::executor::batchingtype::kinflight (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingType9kINFLIGHTE", false]], "tensorrt_llm::executor::batchingtype::kstatic (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12BatchingType7kSTATICE", false]], "tensorrt_llm::executor::beamtokens (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10BeamTokensE", false]], "tensorrt_llm::executor::bufferview (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10BufferViewE", false]], "tensorrt_llm::executor::cachetransceiverconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfigE", false]], "tensorrt_llm::executor::cachetransceiverconfig::cachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", false]], "tensorrt_llm::executor::cachetransceiverconfig::getmaxnumtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfig15getMaxNumTokensEv", false]], "tensorrt_llm::executor::cachetransceiverconfig::mmaxnumtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig13mMaxNumTokensE", false]], "tensorrt_llm::executor::cachetransceiverconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", false]], "tensorrt_llm::executor::cachetransceiverconfig::setmaxnumtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", false]], "tensorrt_llm::executor::capacityschedulerpolicy (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicyE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kguaranteed_no_evict (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy20kGUARANTEED_NO_EVICTE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kmax_utilization (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy16kMAX_UTILIZATIONE", false]], "tensorrt_llm::executor::capacityschedulerpolicy::kstatic_batch (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy13kSTATIC_BATCHE", false]], "tensorrt_llm::executor::communicationmode (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationModeE", false]], "tensorrt_llm::executor::communicationmode::kleader (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationMode7kLEADERE", false]], "tensorrt_llm::executor::communicationmode::korchestrator (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationMode13kORCHESTRATORE", false]], "tensorrt_llm::executor::communicationtype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationTypeE", false]], "tensorrt_llm::executor::communicationtype::kmpi (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor17CommunicationType4kMPIE", false]], "tensorrt_llm::executor::contextchunkingpolicy (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicyE", false]], "tensorrt_llm::executor::contextchunkingpolicy::kequal_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy15kEQUAL_PROGRESSE", false]], "tensorrt_llm::executor::contextchunkingpolicy::kfirst_come_first_served (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy24kFIRST_COME_FIRST_SERVEDE", false]], "tensorrt_llm::executor::contextphaseparams (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsE", false]], "tensorrt_llm::executor::contextphaseparams::contextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERR18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::deleter (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", false]], "tensorrt_llm::executor::contextphaseparams::getdrafttokens (c++ function)": [[0, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams14getDraftTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::getfirstgentokens (c++ function)": [[0, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams17getFirstGenTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::getreqid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getReqIdEv", false]], "tensorrt_llm::executor::contextphaseparams::getserializedstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams18getSerializedStateEv", false]], "tensorrt_llm::executor::contextphaseparams::getstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8getStateEv", false], [0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getStateEv", false]], "tensorrt_llm::executor::contextphaseparams::mdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12mDraftTokensE", false]], "tensorrt_llm::executor::contextphaseparams::mfirstgentokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams15mFirstGenTokensE", false]], "tensorrt_llm::executor::contextphaseparams::mreqid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mReqIdE", false]], "tensorrt_llm::executor::contextphaseparams::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mStateE", false]], "tensorrt_llm::executor::contextphaseparams::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERR18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParamseqERK18ContextPhaseParams", false]], "tensorrt_llm::executor::contextphaseparams::popfirstgentokens (c++ function)": [[0, "_CPPv4NO12tensorrt_llm8executor18ContextPhaseParams17popFirstGenTokensEv", false]], "tensorrt_llm::executor::contextphaseparams::releasestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12releaseStateEv", false]], "tensorrt_llm::executor::contextphaseparams::requestidtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams13RequestIdTypeE", false]], "tensorrt_llm::executor::contextphaseparams::stateptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8StatePtrE", false]], "tensorrt_llm::executor::contextphaseparams::~contextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsD0Ev", false]], "tensorrt_llm::executor::datatransceiverstate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverStateE", false]], "tensorrt_llm::executor::datatransceiverstate::datatransceiverstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", false], [0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::getcachestate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState13getCacheStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::getcommstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState12getCommStateEv", false]], "tensorrt_llm::executor::datatransceiverstate::mcachestate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState11mCacheStateE", false]], "tensorrt_llm::executor::datatransceiverstate::mcommstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState10mCommStateE", false]], "tensorrt_llm::executor::datatransceiverstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", false]], "tensorrt_llm::executor::datatransceiverstate::setcachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::datatransceiverstate::setcommstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", false]], "tensorrt_llm::executor::datatransceiverstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState8toStringEv", false]], "tensorrt_llm::executor::datatype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8DataTypeE", false]], "tensorrt_llm::executor::datatype::kbf16 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kBF16E", false]], "tensorrt_llm::executor::datatype::kbool (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kBOOLE", false]], "tensorrt_llm::executor::datatype::kfp16 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kFP16E", false]], "tensorrt_llm::executor::datatype::kfp32 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kFP32E", false]], "tensorrt_llm::executor::datatype::kfp8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType4kFP8E", false]], "tensorrt_llm::executor::datatype::kint32 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kINT32E", false]], "tensorrt_llm::executor::datatype::kint64 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kINT64E", false]], "tensorrt_llm::executor::datatype::kint8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType5kINT8E", false]], "tensorrt_llm::executor::datatype::kuint8 (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType6kUINT8E", false]], "tensorrt_llm::executor::datatype::kunknown (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8DataType8kUNKNOWNE", false]], "tensorrt_llm::executor::debugconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfigE", false]], "tensorrt_llm::executor::debugconfig::debugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", false]], "tensorrt_llm::executor::debugconfig::getdebuginputtensors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig20getDebugInputTensorsEv", false]], "tensorrt_llm::executor::debugconfig::getdebugoutputtensors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig21getDebugOutputTensorsEv", false]], "tensorrt_llm::executor::debugconfig::getdebugtensornames (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig19getDebugTensorNamesEv", false]], "tensorrt_llm::executor::debugconfig::getdebugtensorsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfig28getDebugTensorsMaxIterationsEv", false]], "tensorrt_llm::executor::debugconfig::mdebuginputtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig18mDebugInputTensorsE", false]], "tensorrt_llm::executor::debugconfig::mdebugoutputtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig19mDebugOutputTensorsE", false]], "tensorrt_llm::executor::debugconfig::mdebugtensornames (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig17mDebugTensorNamesE", false]], "tensorrt_llm::executor::debugconfig::mdebugtensorsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig26mDebugTensorsMaxIterationsE", false]], "tensorrt_llm::executor::debugconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", false]], "tensorrt_llm::executor::debugconfig::setdebuginputtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", false]], "tensorrt_llm::executor::debugconfig::setdebugoutputtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", false]], "tensorrt_llm::executor::debugconfig::setdebugtensornames (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", false]], "tensorrt_llm::executor::debugconfig::setdebugtensorsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::debugconfig::stringvec (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11DebugConfig9StringVecE", false]], "tensorrt_llm::executor::debugtensorsperiteration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIterationE", false]], "tensorrt_llm::executor::debugtensorsperiteration::debugtensors (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration12debugTensorsE", false]], "tensorrt_llm::executor::debugtensorsperiteration::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration4iterE", false]], "tensorrt_llm::executor::decodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfigE", false]], "tensorrt_llm::executor::decodingconfig::decodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", false]], "tensorrt_llm::executor::decodingconfig::enableseamlesslookaheaddecoding (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31enableSeamlessLookaheadDecodingEv", false]], "tensorrt_llm::executor::decodingconfig::getdecodingmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig15getDecodingModeEv", false]], "tensorrt_llm::executor::decodingconfig::geteagleconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig14getEagleConfigEv", false]], "tensorrt_llm::executor::decodingconfig::getlookaheaddecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig26getLookaheadDecodingConfigEv", false]], "tensorrt_llm::executor::decodingconfig::getlookaheaddecodingmaxnumrequest (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig33getLookaheadDecodingMaxNumRequestEv", false]], "tensorrt_llm::executor::decodingconfig::getmedusachoices (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig16getMedusaChoicesEv", false]], "tensorrt_llm::executor::decodingconfig::mdecodingmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig13mDecodingModeE", false]], "tensorrt_llm::executor::decodingconfig::meagleconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig12mEagleConfigE", false]], "tensorrt_llm::executor::decodingconfig::mlookaheaddecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig24mLookaheadDecodingConfigE", false]], "tensorrt_llm::executor::decodingconfig::mlookaheaddecodingmaxnumrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31mLookaheadDecodingMaxNumRequestE", false]], "tensorrt_llm::executor::decodingconfig::mmedusachoices (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14mMedusaChoicesE", false]], "tensorrt_llm::executor::decodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", false]], "tensorrt_llm::executor::decodingconfig::setdecodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig15setDecodingModeERK12DecodingMode", false]], "tensorrt_llm::executor::decodingconfig::seteagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14setEagleConfigERK11EagleConfig", false]], "tensorrt_llm::executor::decodingconfig::setlookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::decodingconfig::setmedusachoices (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14DecodingConfig16setMedusaChoicesERK13MedusaChoices", false]], "tensorrt_llm::executor::decodingmode (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingModeE", false]], "tensorrt_llm::executor::decodingmode::allbitset (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::anybitset (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::auto (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4AutoEv", false]], "tensorrt_llm::executor::decodingmode::beamsearch (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode10BeamSearchEv", false]], "tensorrt_llm::executor::decodingmode::decodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", false]], "tensorrt_llm::executor::decodingmode::eagle (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5EagleEv", false]], "tensorrt_llm::executor::decodingmode::explicitdrafttokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExplicitDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::externaldrafttokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExternalDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::getname (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7getNameEv", false]], "tensorrt_llm::executor::decodingmode::getstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8getStateEv", false]], "tensorrt_llm::executor::decodingmode::isauto (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isAutoEv", false]], "tensorrt_llm::executor::decodingmode::isbeamsearch (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isBeamSearchEv", false]], "tensorrt_llm::executor::decodingmode::iseagle (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7isEagleEv", false]], "tensorrt_llm::executor::decodingmode::isexplicitdrafttokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExplicitDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::isexternaldrafttokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExternalDraftTokensEv", false]], "tensorrt_llm::executor::decodingmode::islookahead (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode11isLookaheadEv", false]], "tensorrt_llm::executor::decodingmode::ismedusa (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8isMedusaEv", false]], "tensorrt_llm::executor::decodingmode::istopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopKEv", false]], "tensorrt_llm::executor::decodingmode::istopkandtopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isTopKandTopPEv", false]], "tensorrt_llm::executor::decodingmode::istopkortopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isTopKorTopPEv", false]], "tensorrt_llm::executor::decodingmode::istopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopPEv", false]], "tensorrt_llm::executor::decodingmode::isusebantokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseBanTokensEv", false]], "tensorrt_llm::executor::decodingmode::isusebanwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isUseBanWordsEv", false]], "tensorrt_llm::executor::decodingmode::isuseexpliciteosstop (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUseExplicitEosStopEv", false]], "tensorrt_llm::executor::decodingmode::isusefrequencypenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isUseFrequencyPenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusemaxlengthstop (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode18isUseMaxLengthStopEv", false]], "tensorrt_llm::executor::decodingmode::isuseminlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseMinLengthEv", false]], "tensorrt_llm::executor::decodingmode::isuseminp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9isUseMinPEv", false]], "tensorrt_llm::executor::decodingmode::isusenorepeatngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseNoRepeatNgramSizeEv", false]], "tensorrt_llm::executor::decodingmode::isuseoccurrencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseOccurrencePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isUsePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusepresencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUsePresencePenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isuserepetitionpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseRepetitionPenaltyEv", false]], "tensorrt_llm::executor::decodingmode::isusestopcriteria (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode17isUseStopCriteriaEv", false]], "tensorrt_llm::executor::decodingmode::isusestopwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseStopWordsEv", false]], "tensorrt_llm::executor::decodingmode::isusetemperature (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode16isUseTemperatureEv", false]], "tensorrt_llm::executor::decodingmode::isusevariablebeamwidthsearch (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingMode28isUseVariableBeamWidthSearchEv", false]], "tensorrt_llm::executor::decodingmode::kauto (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kAutoE", false]], "tensorrt_llm::executor::decodingmode::kbeamsearch (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode11kBeamSearchE", false]], "tensorrt_llm::executor::decodingmode::keagle (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6kEagleE", false]], "tensorrt_llm::executor::decodingmode::kexplicitdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExplicitDraftTokensE", false]], "tensorrt_llm::executor::decodingmode::kexternaldrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExternalDraftTokensE", false]], "tensorrt_llm::executor::decodingmode::klookahead (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode10kLookaheadE", false]], "tensorrt_llm::executor::decodingmode::kmedusa (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode7kMedusaE", false]], "tensorrt_llm::executor::decodingmode::knumflags (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kNumFlagsE", false]], "tensorrt_llm::executor::decodingmode::ktopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopKE", false]], "tensorrt_llm::executor::decodingmode::ktopktopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kTopKTopPE", false]], "tensorrt_llm::executor::decodingmode::ktopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopPE", false]], "tensorrt_llm::executor::decodingmode::kusebantokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseBanTokensE", false]], "tensorrt_llm::executor::decodingmode::kusebanwords (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12kUseBanWordsE", false]], "tensorrt_llm::executor::decodingmode::kuseexpliciteosstop (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19kUseExplicitEosStopE", false]], "tensorrt_llm::executor::decodingmode::kusefrequencypenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode22kUseFrequencyPenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusemaxlengthstop (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode17kUseMaxLengthStopE", false]], "tensorrt_llm::executor::decodingmode::kuseminlength (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseMinLengthE", false]], "tensorrt_llm::executor::decodingmode::kuseminp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8kUseMinPE", false]], "tensorrt_llm::executor::decodingmode::kusenorepeatngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUseNoRepeatNgramSizeE", false]], "tensorrt_llm::executor::decodingmode::kuseoccurrencepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseOccurrencePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUsePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusepresencepenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUsePresencePenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kuserepetitionpenalties (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseRepetitionPenaltiesE", false]], "tensorrt_llm::executor::decodingmode::kusestandardstopcriteria (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode24kUseStandardStopCriteriaE", false]], "tensorrt_llm::executor::decodingmode::kusestopwords (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseStopWordsE", false]], "tensorrt_llm::executor::decodingmode::kusetemperature (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode15kUseTemperatureE", false]], "tensorrt_llm::executor::decodingmode::kusevariablebeamwidthsearch (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode27kUseVariableBeamWidthSearchE", false]], "tensorrt_llm::executor::decodingmode::lookahead (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode9LookaheadEv", false]], "tensorrt_llm::executor::decodingmode::medusa (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6MedusaEv", false]], "tensorrt_llm::executor::decodingmode::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode6mStateE", false]], "tensorrt_llm::executor::decodingmode::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", false]], "tensorrt_llm::executor::decodingmode::setbitto (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", false]], "tensorrt_llm::executor::decodingmode::topk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopKEv", false]], "tensorrt_llm::executor::decodingmode::topktopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode8TopKTopPEv", false]], "tensorrt_llm::executor::decodingmode::topp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopPEv", false]], "tensorrt_llm::executor::decodingmode::underlyingtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode14UnderlyingTypeE", false]], "tensorrt_llm::executor::decodingmode::usebantokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", false]], "tensorrt_llm::executor::decodingmode::usebanwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", false]], "tensorrt_llm::executor::decodingmode::useexpliciteosstop (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", false]], "tensorrt_llm::executor::decodingmode::usefrequencypenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", false]], "tensorrt_llm::executor::decodingmode::usemaxlengthstop (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", false]], "tensorrt_llm::executor::decodingmode::useminlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", false]], "tensorrt_llm::executor::decodingmode::useminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", false]], "tensorrt_llm::executor::decodingmode::usenorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", false]], "tensorrt_llm::executor::decodingmode::useoccurrencepenalties (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", false]], "tensorrt_llm::executor::decodingmode::usepresencepenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", false]], "tensorrt_llm::executor::decodingmode::userepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", false]], "tensorrt_llm::executor::decodingmode::usestopwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", false]], "tensorrt_llm::executor::decodingmode::usetemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", false]], "tensorrt_llm::executor::decodingmode::usevariablebeamwidthsearch (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", false]], "tensorrt_llm::executor::detail (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6detailE", false]], "tensorrt_llm::executor::detail::dimtype64 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9DimType64E", false]], "tensorrt_llm::executor::detail::ofitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", false]], "tensorrt_llm::executor::detail::toitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", false]], "tensorrt_llm::executor::disagg_executor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executorE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::awaitcontextresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::awaitgenerationresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::canenqueue (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator10canEnqueueEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::disaggexecutororchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::enqueuecontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::enqueuegeneration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::getcontextexecutors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator19getContextExecutorsEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::getgenexecutors (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator15getGenExecutorsEv", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator5mImplE", false]], "tensorrt_llm::executor::disagg_executor::disaggexecutororchestrator::~disaggexecutororchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorD0Ev", false]], "tensorrt_llm::executor::disagg_executor::responsewithid (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::gid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId3gidE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::response (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId8responseE", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::responsewithid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", false], [0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", false]], "tensorrt_llm::executor::disagg_executor::responsewithid::~responsewithid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdD0Ev", false]], "tensorrt_llm::executor::disservingrequeststats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStatsE", false]], "tensorrt_llm::executor::disservingrequeststats::kvcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats11kvCacheSizeE", false]], "tensorrt_llm::executor::disservingrequeststats::kvcachetransferms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats17kvCacheTransferMSE", false]], "tensorrt_llm::executor::dynamicbatchconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfigE", false]], "tensorrt_llm::executor::dynamicbatchconfig::dynamicbatchconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", false]], "tensorrt_llm::executor::dynamicbatchconfig::getbatchsizetable (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig17getBatchSizeTableEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getdynamicbatchmovingaveragewindow (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig34getDynamicBatchMovingAverageWindowEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getenablebatchsizetuning (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig24getEnableBatchSizeTuningEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::getenablemaxnumtokenstuning (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig27getEnableMaxNumTokensTuningEv", false]], "tensorrt_llm::executor::dynamicbatchconfig::kdefaultbatchsizetable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22kDefaultBatchSizeTableE", false]], "tensorrt_llm::executor::dynamicbatchconfig::kdefaultdynamicbatchmovingaveragewindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig39kDefaultDynamicBatchMovingAverageWindowE", false]], "tensorrt_llm::executor::dynamicbatchconfig::mbatchsizetable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig15mBatchSizeTableE", false]], "tensorrt_llm::executor::dynamicbatchconfig::mdynamicbatchmovingaveragewindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig32mDynamicBatchMovingAverageWindowE", false]], "tensorrt_llm::executor::dynamicbatchconfig::menablebatchsizetuning (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22mEnableBatchSizeTuningE", false]], "tensorrt_llm::executor::dynamicbatchconfig::menablemaxnumtokenstuning (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig25mEnableMaxNumTokensTuningE", false]], "tensorrt_llm::executor::eaglechoices (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12EagleChoicesE", false]], "tensorrt_llm::executor::eagleconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfigE", false]], "tensorrt_llm::executor::eagleconfig::checkposteriorvalue (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", false]], "tensorrt_llm::executor::eagleconfig::eagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::eagleconfig::getdynamictreemaxtopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getDynamicTreeMaxTopKEv", false]], "tensorrt_llm::executor::eagleconfig::geteaglechoices (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig15getEagleChoicesEv", false]], "tensorrt_llm::executor::eagleconfig::getposteriorthreshold (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getPosteriorThresholdEv", false]], "tensorrt_llm::executor::eagleconfig::isgreedysampling (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig16isGreedySamplingEv", false]], "tensorrt_llm::executor::eagleconfig::mdynamictreemaxtopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mDynamicTreeMaxTopKE", false]], "tensorrt_llm::executor::eagleconfig::meaglechoices (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig13mEagleChoicesE", false]], "tensorrt_llm::executor::eagleconfig::mgreedysampling (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mGreedySamplingE", false]], "tensorrt_llm::executor::eagleconfig::mposteriorthreshold (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mPosteriorThresholdE", false]], "tensorrt_llm::executor::eagleconfig::musedynamictree (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mUseDynamicTreeE", false]], "tensorrt_llm::executor::eagleconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", false]], "tensorrt_llm::executor::eagleconfig::usedynamictree (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11EagleConfig14useDynamicTreeEv", false]], "tensorrt_llm::executor::executor (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutorE", false]], "tensorrt_llm::executor::executor::awaitresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::executor::cancelrequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", false]], "tensorrt_llm::executor::executor::canenqueuerequests (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor18canEnqueueRequestsEv", false]], "tensorrt_llm::executor::executor::enqueuerequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", false]], "tensorrt_llm::executor::executor::enqueuerequests (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", false]], "tensorrt_llm::executor::executor::executor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERR8Executor", false]], "tensorrt_llm::executor::executor::getkvcacheeventmanager (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor22getKVCacheEventManagerEv", false]], "tensorrt_llm::executor::executor::getlatestdebugtensors (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestDebugTensorsEv", false]], "tensorrt_llm::executor::executor::getlatestiterationstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor23getLatestIterationStatsEv", false]], "tensorrt_llm::executor::executor::getlatestrequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestRequestStatsEv", false]], "tensorrt_llm::executor::executor::getnumresponsesready (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", false]], "tensorrt_llm::executor::executor::isparticipant (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Executor13isParticipantEv", false]], "tensorrt_llm::executor::executor::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor5mImplE", false]], "tensorrt_llm::executor::executor::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", false], [0, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERR8Executor", false]], "tensorrt_llm::executor::executor::shutdown (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Executor8shutdownEv", false]], "tensorrt_llm::executor::executor::~executor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ExecutorD0Ev", false]], "tensorrt_llm::executor::executorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfigE", false]], "tensorrt_llm::executor::executorconfig::executorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", false]], "tensorrt_llm::executor::executorconfig::getadditionalmodeloutputs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getAdditionalModelOutputsEv", false]], "tensorrt_llm::executor::executorconfig::getbatchingtype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getBatchingTypeEv", false]], "tensorrt_llm::executor::executorconfig::getcachetransceiverconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getCacheTransceiverConfigEv", false]], "tensorrt_llm::executor::executorconfig::getdebugconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig14getDebugConfigEv", false]], "tensorrt_llm::executor::executorconfig::getdecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getDecodingConfigEv", false]], "tensorrt_llm::executor::executorconfig::getenablechunkedcontext (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getEnableChunkedContextEv", false]], "tensorrt_llm::executor::executorconfig::getenabletrtoverlap (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getEnableTrtOverlapEv", false]], "tensorrt_llm::executor::executorconfig::getextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig32getExtendedRuntimePerfKnobConfigEv", false]], "tensorrt_llm::executor::executorconfig::getgathergenerationlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getGatherGenerationLogitsEv", false]], "tensorrt_llm::executor::executorconfig::getgpuweightspercent (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getGpuWeightsPercentEv", false]], "tensorrt_llm::executor::executorconfig::getguideddecodingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getGuidedDecodingConfigEv", false]], "tensorrt_llm::executor::executorconfig::getiterstatsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getIterStatsMaxIterationsEv", false]], "tensorrt_llm::executor::executorconfig::getkvcacheconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getKvCacheConfigEv", false]], "tensorrt_llm::executor::executorconfig::getkvcacheconfigref (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19getKvCacheConfigRefEv", false]], "tensorrt_llm::executor::executorconfig::getlogitspostprocessorconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getLogitsPostProcessorConfigEv", false]], "tensorrt_llm::executor::executorconfig::getmaxbatchsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBatchSizeEv", false]], "tensorrt_llm::executor::executorconfig::getmaxbeamwidth (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::executor::executorconfig::getmaxnumtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxNumTokensEv", false]], "tensorrt_llm::executor::executorconfig::getmaxqueuesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxQueueSizeEv", false]], "tensorrt_llm::executor::executorconfig::getmaxseqidlemicroseconds (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getMaxSeqIdleMicrosecondsEv", false]], "tensorrt_llm::executor::executorconfig::getnormalizelogprobs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getNormalizeLogProbsEv", false]], "tensorrt_llm::executor::executorconfig::getparallelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getParallelConfigEv", false]], "tensorrt_llm::executor::executorconfig::getpeftcacheconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getPeftCacheConfigEv", false]], "tensorrt_llm::executor::executorconfig::getprompttableoffloading (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig24getPromptTableOffloadingEv", false]], "tensorrt_llm::executor::executorconfig::getrecvpollperiodms (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getRecvPollPeriodMsEv", false]], "tensorrt_llm::executor::executorconfig::getrequeststatsmaxiterations (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getRequestStatsMaxIterationsEv", false]], "tensorrt_llm::executor::executorconfig::getschedulerconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getSchedulerConfigEv", false]], "tensorrt_llm::executor::executorconfig::getschedulerconfigref (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21getSchedulerConfigRefEv", false]], "tensorrt_llm::executor::executorconfig::getspecdecconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getSpecDecConfigEv", false]], "tensorrt_llm::executor::executorconfig::getusegpudirectstorage (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig22getUseGpuDirectStorageEv", false]], "tensorrt_llm::executor::executorconfig::kdefaultiterstatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultIterStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::kdefaultmaxseqidlemicroseconds (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultMaxSeqIdleMicrosecondsE", false]], "tensorrt_llm::executor::executorconfig::kdefaultrequeststatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig33kDefaultRequestStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::madditionalmodeloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mAdditionalModelOutputsE", false]], "tensorrt_llm::executor::executorconfig::mbatchingtype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mBatchingTypeE", false]], "tensorrt_llm::executor::executorconfig::mcachetransceiverconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mCacheTransceiverConfigE", false]], "tensorrt_llm::executor::executorconfig::mdebugconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig12mDebugConfigE", false]], "tensorrt_llm::executor::executorconfig::mdecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::menablechunkedcontext (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mEnableChunkedContextE", false]], "tensorrt_llm::executor::executorconfig::menabletrtoverlap (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mEnableTrtOverlapE", false]], "tensorrt_llm::executor::executorconfig::mextendedruntimeperfknobconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30mExtendedRuntimePerfKnobConfigE", false]], "tensorrt_llm::executor::executorconfig::mgathergenerationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mGatherGenerationLogitsE", false]], "tensorrt_llm::executor::executorconfig::mgpuweightspercent (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mGpuWeightsPercentE", false]], "tensorrt_llm::executor::executorconfig::mguideddecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mGuidedDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::miterstatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mIterStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::mkvcacheconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14mKvCacheConfigE", false]], "tensorrt_llm::executor::executorconfig::mlogitspostprocessorconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mLogitsPostProcessorConfigE", false]], "tensorrt_llm::executor::executorconfig::mmaxbatchsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBatchSizeE", false]], "tensorrt_llm::executor::executorconfig::mmaxbeamwidth (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBeamWidthE", false]], "tensorrt_llm::executor::executorconfig::mmaxnumtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxNumTokensE", false]], "tensorrt_llm::executor::executorconfig::mmaxqueuesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxQueueSizeE", false]], "tensorrt_llm::executor::executorconfig::mmaxseqidlemicroseconds (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mMaxSeqIdleMicrosecondsE", false]], "tensorrt_llm::executor::executorconfig::mnormalizelogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mNormalizeLogProbsE", false]], "tensorrt_llm::executor::executorconfig::mparallelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mParallelConfigE", false]], "tensorrt_llm::executor::executorconfig::mpeftcacheconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mPeftCacheConfigE", false]], "tensorrt_llm::executor::executorconfig::mprompttableoffloading (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22mPromptTableOffloadingE", false]], "tensorrt_llm::executor::executorconfig::mrecvpollperiodms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mRecvPollPeriodMsE", false]], "tensorrt_llm::executor::executorconfig::mrequeststatsmaxiterations (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mRequestStatsMaxIterationsE", false]], "tensorrt_llm::executor::executorconfig::mschedulerconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mSchedulerConfigE", false]], "tensorrt_llm::executor::executorconfig::mspeculativedecodingconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mSpeculativeDecodingConfigE", false]], "tensorrt_llm::executor::executorconfig::musegpudirectstorage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20mUseGpuDirectStorageE", false]], "tensorrt_llm::executor::executorconfig::setadditionalmodeloutputs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", false]], "tensorrt_llm::executor::executorconfig::setbatchingtype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", false]], "tensorrt_llm::executor::executorconfig::setcachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", false]], "tensorrt_llm::executor::executorconfig::setdebugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", false]], "tensorrt_llm::executor::executorconfig::setdecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setenablechunkedcontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", false]], "tensorrt_llm::executor::executorconfig::setenabletrtoverlap (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", false]], "tensorrt_llm::executor::executorconfig::setextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", false]], "tensorrt_llm::executor::executorconfig::setgathergenerationlogits (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", false]], "tensorrt_llm::executor::executorconfig::setgpuweightspercent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", false]], "tensorrt_llm::executor::executorconfig::setguideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setiterstatsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setkvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", false]], "tensorrt_llm::executor::executorconfig::setlogitspostprocessorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", false]], "tensorrt_llm::executor::executorconfig::setmaxbatchsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxnumtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setmaxqueuesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::executorconfig::setmaxseqidlemicroseconds (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", false]], "tensorrt_llm::executor::executorconfig::setnormalizelogprobs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", false]], "tensorrt_llm::executor::executorconfig::setparallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", false]], "tensorrt_llm::executor::executorconfig::setpeftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", false]], "tensorrt_llm::executor::executorconfig::setprompttableoffloading (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", false]], "tensorrt_llm::executor::executorconfig::setrecvpollperiodms (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setrequeststatsmaxiterations (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", false]], "tensorrt_llm::executor::executorconfig::setschedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", false]], "tensorrt_llm::executor::executorconfig::setspecdecconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", false]], "tensorrt_llm::executor::executorconfig::setusegpudirectstorage (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::extendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getcudagraphcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21getCudaGraphCacheSizeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getcudagraphmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16getCudaGraphModeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getenablecontextfmhafp32acc (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27getEnableContextFMHAFP32AccEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::getmultiblockmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17getMultiBlockModeEv", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mcudagraphcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig19mCudaGraphCacheSizeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mcudagraphmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig14mCudaGraphModeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::menablecontextfmhafp32acc (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig25mEnableContextFMHAFP32AccE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::mmultiblockmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig15mMultiBlockModeE", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setcudagraphcachesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setcudagraphmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setenablecontextfmhafp32acc (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", false]], "tensorrt_llm::executor::extendedruntimeperfknobconfig::setmultiblockmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", false]], "tensorrt_llm::executor::externaldrafttokensconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfigE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::externaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getacceptancethreshold (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig22getAcceptanceThresholdEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getfastlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig13getFastLogitsEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::getlogits (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getLogitsEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::gettokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getTokensEv", false]], "tensorrt_llm::executor::externaldrafttokensconfig::macceptancethreshold (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig20mAcceptanceThresholdE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mfastlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig11mFastLogitsE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mLogitsE", false]], "tensorrt_llm::executor::externaldrafttokensconfig::mtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mTokensE", false]], "tensorrt_llm::executor::finishreason (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReasonE", false]], "tensorrt_llm::executor::finishreason::kcancelled (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason10kCANCELLEDE", false]], "tensorrt_llm::executor::finishreason::kend_id (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason7kEND_IDE", false]], "tensorrt_llm::executor::finishreason::klength (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason7kLENGTHE", false]], "tensorrt_llm::executor::finishreason::knot_finished (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason13kNOT_FINISHEDE", false]], "tensorrt_llm::executor::finishreason::kstop_words (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason11kSTOP_WORDSE", false]], "tensorrt_llm::executor::finishreason::ktimed_out (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12FinishReason10kTIMED_OUTE", false]], "tensorrt_llm::executor::floattype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9FloatTypeE", false]], "tensorrt_llm::executor::guideddecodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfigE", false]], "tensorrt_llm::executor::guideddecodingconfig::getbackend (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig10getBackendEv", false]], "tensorrt_llm::executor::guideddecodingconfig::getencodedvocab (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getEncodedVocabEv", false]], "tensorrt_llm::executor::guideddecodingconfig::getstoptokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getStopTokenIdsEv", false]], "tensorrt_llm::executor::guideddecodingconfig::gettokenizerstr (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getTokenizerStrEv", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingbackend (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackendE", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingbackend::kxgrammar (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackend9kXGRAMMARE", false]], "tensorrt_llm::executor::guideddecodingconfig::guideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", false]], "tensorrt_llm::executor::guideddecodingconfig::mbackend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig8mBackendE", false]], "tensorrt_llm::executor::guideddecodingconfig::mencodedvocab (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mEncodedVocabE", false]], "tensorrt_llm::executor::guideddecodingconfig::mstoptokenids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mStopTokenIdsE", false]], "tensorrt_llm::executor::guideddecodingconfig::mtokenizerstr (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mTokenizerStrE", false]], "tensorrt_llm::executor::guideddecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", false]], "tensorrt_llm::executor::guideddecodingconfig::setbackend (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", false]], "tensorrt_llm::executor::guideddecodingconfig::setencodedvocab (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", false]], "tensorrt_llm::executor::guideddecodingconfig::setstoptokenids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", false]], "tensorrt_llm::executor::guideddecodingconfig::settokenizerstr (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", false]], "tensorrt_llm::executor::guideddecodingconfig::validate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig8validateEv", false]], "tensorrt_llm::executor::guideddecodingparams (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParamsE", false]], "tensorrt_llm::executor::guideddecodingparams::getguide (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams8getGuideEv", false]], "tensorrt_llm::executor::guideddecodingparams::getguidetype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams12getGuideTypeEv", false]], "tensorrt_llm::executor::guideddecodingparams::guideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideTypeE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kebnf_grammar (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType13kEBNF_GRAMMARE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kjson (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType5kJSONE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kjson_schema (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType12kJSON_SCHEMAE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kregex (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType6kREGEXE", false]], "tensorrt_llm::executor::guideddecodingparams::guidetype::kstructural_tag (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType15kSTRUCTURAL_TAGE", false]], "tensorrt_llm::executor::guideddecodingparams::mguide (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams6mGuideE", false]], "tensorrt_llm::executor::guideddecodingparams::mguidetype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams10mGuideTypeE", false]], "tensorrt_llm::executor::guideddecodingparams::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", false]], "tensorrt_llm::executor::idtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6IdTypeE", false]], "tensorrt_llm::executor::inflightbatchingstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStatsE", false]], "tensorrt_llm::executor::inflightbatchingstats::avgnumdecodedtokensperiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats26avgNumDecodedTokensPerIterE", false]], "tensorrt_llm::executor::inflightbatchingstats::microbatchid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12microBatchIdE", false]], "tensorrt_llm::executor::inflightbatchingstats::numcontextrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats18numContextRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numctxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12numCtxTokensE", false]], "tensorrt_llm::executor::inflightbatchingstats::numgenrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats14numGenRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numpausedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats17numPausedRequestsE", false]], "tensorrt_llm::executor::inflightbatchingstats::numscheduledrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats20numScheduledRequestsE", false]], "tensorrt_llm::executor::iterationstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStatsE", false]], "tensorrt_llm::executor::iterationstats::cpumemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats11cpuMemUsageE", false]], "tensorrt_llm::executor::iterationstats::crosskvcachestats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17crossKvCacheStatsE", false]], "tensorrt_llm::executor::iterationstats::gpumemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats11gpuMemUsageE", false]], "tensorrt_llm::executor::iterationstats::inflightbatchingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats21inflightBatchingStatsE", false]], "tensorrt_llm::executor::iterationstats::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats4iterE", false]], "tensorrt_llm::executor::iterationstats::iterlatencyms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats13iterLatencyMSE", false]], "tensorrt_llm::executor::iterationstats::kvcachestats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats12kvCacheStatsE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizeruntime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxBatchSizeRuntimeE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizestatic (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxBatchSizeStaticE", false]], "tensorrt_llm::executor::iterationstats::maxbatchsizetunerrecommended (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxBatchSizeTunerRecommendedE", false]], "tensorrt_llm::executor::iterationstats::maxnumactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20maxNumActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokensruntime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxNumTokensRuntimeE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokensstatic (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxNumTokensStaticE", false]], "tensorrt_llm::executor::iterationstats::maxnumtokenstunerrecommended (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxNumTokensTunerRecommendedE", false]], "tensorrt_llm::executor::iterationstats::newactiverequestsqueuelatencyms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats31newActiveRequestsQueueLatencyMSE", false]], "tensorrt_llm::executor::iterationstats::numactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17numActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::numcompletedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20numCompletedRequestsE", false]], "tensorrt_llm::executor::iterationstats::numnewactiverequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats20numNewActiveRequestsE", false]], "tensorrt_llm::executor::iterationstats::numqueuedrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats17numQueuedRequestsE", false]], "tensorrt_llm::executor::iterationstats::pinnedmemusage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats14pinnedMemUsageE", false]], "tensorrt_llm::executor::iterationstats::specdecstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats12specDecStatsE", false]], "tensorrt_llm::executor::iterationstats::staticbatchingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats19staticBatchingStatsE", false]], "tensorrt_llm::executor::iterationstats::timestamp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14IterationStats9timestampE", false]], "tensorrt_llm::executor::iterationtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor13IterationTypeE", false]], "tensorrt_llm::executor::jsonserialization (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor17JsonSerializationE", false]], "tensorrt_llm::executor::jsonserialization::tojsonstr (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", false]], "tensorrt_llm::executor::kv_cache (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", false]], "tensorrt_llm::executor::kv_cache::agentdesc (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDescE", false]], "tensorrt_llm::executor::kv_cache::agentdesc::agentdesc (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDesc9AgentDescENSt6stringE", false]], "tensorrt_llm::executor::kv_cache::agentdesc::getbackendagentdesc (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9AgentDesc19getBackendAgentDescEv", false]], "tensorrt_llm::executor::kv_cache::agentdesc::mbackendagentdesc (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDesc17mBackendAgentDescE", false]], "tensorrt_llm::executor::kv_cache::agentstate (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentStateE", false]], "tensorrt_llm::executor::kv_cache::agentstate::agentstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateENSt6stringENSt6stringE", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateEv", false]], "tensorrt_llm::executor::kv_cache::agentstate::magentname (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10mAgentNameE", false]], "tensorrt_llm::executor::kv_cache::agentstate::mconnectioninfo (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState15mConnectionInfoE", false]], "tensorrt_llm::executor::kv_cache::agentstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10AgentStateeqERK10AgentState", false]], "tensorrt_llm::executor::kv_cache::agentstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10AgentState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::baseagentconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfigE", false]], "tensorrt_llm::executor::kv_cache::baseagentconfig::mname (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfig5mNameE", false]], "tensorrt_llm::executor::kv_cache::baseagentconfig::useprogthread (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfig13useProgThreadE", false]], "tensorrt_llm::executor::kv_cache::basetransferagent (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgentE", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::checkremotedescs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16checkRemoteDescsERKNSt6stringERK11MemoryDescs", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::connectremoteagent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent18connectRemoteAgentERKNSt6stringERK18ConnectionInfoType", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::deregistermemory (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16deregisterMemoryERK13RegisterDescs", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::getconnectioninfo (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17getConnectionInfoEv", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::getlocalagentdesc (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17getLocalAgentDescEv", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::getnotifiedsyncmessages (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent23getNotifiedSyncMessagesEv", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::invalidateremoteagent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent21invalidateRemoteAgentERKNSt6stringE", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::loadremoteagent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent15loadRemoteAgentERKNSt6stringERK9AgentDesc", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::notifysyncmessage (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17notifySyncMessageERKNSt6stringERK11SyncMessage", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::registermemory (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent14registerMemoryERK13RegisterDescs", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::submittransferrequests (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent22submitTransferRequestsERK15TransferRequest", false]], "tensorrt_llm::executor::kv_cache::basetransferagent::~basetransferagent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgentD0Ev", false]], "tensorrt_llm::executor::kv_cache::cachestate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::attentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::mattentiontype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig14mAttentionTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentionconfig::mkvfactor (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig9mKvFactorE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype::kdefault (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType8kDEFAULTE", false]], "tensorrt_llm::executor::kv_cache::cachestate::attentiontype::kmla (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType4kMLAE", false]], "tensorrt_llm::executor::kv_cache::cachestate::cachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", false]], "tensorrt_llm::executor::kv_cache::cachestate::getattentionconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState18getAttentionConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getdatatype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11getDataTypeEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getmodelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14getModelConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::getparallelconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState17getParallelConfigEv", false]], "tensorrt_llm::executor::kv_cache::cachestate::mattentionconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState16mAttentionConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::mdatatype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState9mDataTypeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::mmodelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState12mModelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::mnbkvheadsperlayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig18mNbKvHeadsPerLayerE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::msizeperhead (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig12mSizePerHeadE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::mtokensperblock (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig15mTokensPerBlockE", false]], "tensorrt_llm::executor::kv_cache::cachestate::modelconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", false]], "tensorrt_llm::executor::kv_cache::cachestate::mparallelconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15mParallelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mdprank (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPrankE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mdpsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPsizeE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::menableattentiondp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mEnableAttentionDPE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mpipelineparallelism (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig20mPipelineParallelismE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::mtensorparallelism (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mTensorParallelismE", false]], "tensorrt_llm::executor::kv_cache::cachestate::parallelconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", false]], "tensorrt_llm::executor::kv_cache::cachestate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::commstate (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommStateE", false]], "tensorrt_llm::executor::kv_cache::commstate::commstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10AgentStateEEi", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getagentstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13getAgentStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getmpistate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState11getMpiStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getselfidx (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10getSelfIdxEv", false]], "tensorrt_llm::executor::kv_cache::commstate::getsocketstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState14getSocketStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::isagentstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState12isAgentStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::ismpistate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10isMpiStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::issocketstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13isSocketStateEv", false]], "tensorrt_llm::executor::kv_cache::commstate::mselfidx (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState8mSelfIdxE", false]], "tensorrt_llm::executor::kv_cache::commstate::mstate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState6mStateE", false]], "tensorrt_llm::executor::kv_cache::commstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", false]], "tensorrt_llm::executor::kv_cache::commstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::connection (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionE", false]], "tensorrt_llm::executor::kv_cache::connection::isthreadsafe (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection12isThreadSafeEv", false]], "tensorrt_llm::executor::kv_cache::connection::recv (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", false]], "tensorrt_llm::executor::kv_cache::connection::send (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", false]], "tensorrt_llm::executor::kv_cache::connection::~connection (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionD0Ev", false]], "tensorrt_llm::executor::kv_cache::connectioninfotype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache18ConnectionInfoTypeE", false]], "tensorrt_llm::executor::kv_cache::connectionmanager (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerE", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::getcommstate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache17ConnectionManager12getCommStateEv", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::getconnections (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::recvconnect (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", false]], "tensorrt_llm::executor::kv_cache::connectionmanager::~connectionmanager (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerD0Ev", false]], "tensorrt_llm::executor::kv_cache::datacontext (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContextE", false]], "tensorrt_llm::executor::kv_cache::datacontext::datacontext (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", false]], "tensorrt_llm::executor::kv_cache::datacontext::gettag (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11DataContext6getTagEv", false]], "tensorrt_llm::executor::kv_cache::datacontext::mtag (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext4mTagE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::dlsym (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader5dlSymEPvPKc", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::dynlibloader (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader12DynLibLoaderERK12DynLibLoader", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader12DynLibLoaderEv", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::getfunctionpointer (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor8kv_cache12DynLibLoader18getFunctionPointerE9FunctionTRKNSt6stringERKNSt6stringE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::gethandle (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9getHandleERKNSt6stringE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::getinstance (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader11getInstanceEv", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::mdllmutex (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9mDllMutexE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::mhandlers (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9mHandlersE", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderaSERK12DynLibLoader", false]], "tensorrt_llm::executor::kv_cache::dynlibloader::~dynlibloader (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderD0Ev", false]], "tensorrt_llm::executor::kv_cache::maketransferagent (c++ function)": [[0, "_CPPv4IDpEN12tensorrt_llm8executor8kv_cache17makeTransferAgentENSt10unique_ptrI17BaseTransferAgentEERKNSt6stringEDpRR4Args", false]], "tensorrt_llm::executor::kv_cache::memorydesc (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDescE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::deserialize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc11deserializeERNSt7istreamE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::getaddr (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc7getAddrEv", false]], "tensorrt_llm::executor::kv_cache::memorydesc::getdeviceid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc11getDeviceIdEv", false]], "tensorrt_llm::executor::kv_cache::memorydesc::getlen (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc6getLenEv", false]], "tensorrt_llm::executor::kv_cache::memorydesc::maddr (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc5mAddrE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::mdeviceid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9mDeviceIdE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::memorydesc (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescE9uintptr_t6size_t8uint32_t", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescEPv6size_t8uint32_t", false], [0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescERKNSt6vectorIcEE8uint32_t", false]], "tensorrt_llm::executor::kv_cache::memorydesc::mlen (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc4mLenE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::serialize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9serializeERK10MemoryDescRNSt7ostreamE", false]], "tensorrt_llm::executor::kv_cache::memorydesc::serializedsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc14serializedSizeERK10MemoryDesc", false]], "tensorrt_llm::executor::kv_cache::memorydescs (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescsE", false]], "tensorrt_llm::executor::kv_cache::memorydescs::getdescs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11MemoryDescs8getDescsEv", false]], "tensorrt_llm::executor::kv_cache::memorydescs::gettype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11MemoryDescs7getTypeEv", false]], "tensorrt_llm::executor::kv_cache::memorydescs::mdescs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs6mDescsE", false]], "tensorrt_llm::executor::kv_cache::memorydescs::memorydescs (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs11MemoryDescsE10MemoryTypeNSt6vectorI10MemoryDescEE", false]], "tensorrt_llm::executor::kv_cache::memorydescs::mtype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs5mTypeE", false]], "tensorrt_llm::executor::kv_cache::memorytype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryTypeE", false]], "tensorrt_llm::executor::kv_cache::memorytype::kblk (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType4kBLKE", false]], "tensorrt_llm::executor::kv_cache::memorytype::kdram (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kDRAME", false]], "tensorrt_llm::executor::kv_cache::memorytype::kfile (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kFILEE", false]], "tensorrt_llm::executor::kv_cache::memorytype::kobj (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType4kOBJE", false]], "tensorrt_llm::executor::kv_cache::memorytype::kvram (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kVRAME", false]], "tensorrt_llm::executor::kv_cache::mpistate (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiStateE", false]], "tensorrt_llm::executor::kv_cache::mpistate::mranks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiState6mRanksE", false]], "tensorrt_llm::executor::kv_cache::mpistate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", false]], "tensorrt_llm::executor::kv_cache::mpistate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::registerdescs (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache13RegisterDescsE", false]], "tensorrt_llm::executor::kv_cache::socketstate (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketStateE", false]], "tensorrt_llm::executor::kv_cache::socketstate::mip (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState3mIpE", false]], "tensorrt_llm::executor::kv_cache::socketstate::mport (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState5mPortE", false]], "tensorrt_llm::executor::kv_cache::socketstate::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", false]], "tensorrt_llm::executor::kv_cache::socketstate::tostring (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketState8toStringEv", false]], "tensorrt_llm::executor::kv_cache::syncmessage (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache11SyncMessageE", false]], "tensorrt_llm::executor::kv_cache::transferdescs (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache13TransferDescsE", false]], "tensorrt_llm::executor::kv_cache::transferop (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOpE", false]], "tensorrt_llm::executor::kv_cache::transferop::kread (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOp5kREADE", false]], "tensorrt_llm::executor::kv_cache::transferop::kwrite (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOp6kWRITEE", false]], "tensorrt_llm::executor::kv_cache::transferrequest (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequestE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::getdstdescs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest11getDstDescsEv", false]], "tensorrt_llm::executor::kv_cache::transferrequest::getop (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest5getOpEv", false]], "tensorrt_llm::executor::kv_cache::transferrequest::getremotename (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest13getRemoteNameEv", false]], "tensorrt_llm::executor::kv_cache::transferrequest::getsrcdescs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest11getSrcDescsEv", false]], "tensorrt_llm::executor::kv_cache::transferrequest::getsyncmessage (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest14getSyncMessageEv", false]], "tensorrt_llm::executor::kv_cache::transferrequest::mdstdescs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest9mDstDescsE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::mop (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest3mOpE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::mremotename (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest11mRemoteNameE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::msrcdescs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest9mSrcDescsE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::msyncmessage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest12mSyncMessageE", false]], "tensorrt_llm::executor::kv_cache::transferrequest::transferrequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", false]], "tensorrt_llm::executor::kv_cache::transferstatus (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache14TransferStatusE", false]], "tensorrt_llm::executor::kv_cache::transferstatus::iscompleted (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache14TransferStatus11isCompletedEv", false]], "tensorrt_llm::executor::kv_cache::transferstatus::wait (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8kv_cache14TransferStatus4waitEv", false]], "tensorrt_llm::executor::kv_cache::transferstatus::~transferstatus (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8kv_cache14TransferStatusD0Ev", false]], "tensorrt_llm::executor::kvcacheconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfigE", false]], "tensorrt_llm::executor::kvcacheconfig::fillemptyfieldsfromruntimedefaults (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", false]], "tensorrt_llm::executor::kvcacheconfig::getcopyonpartialreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getCopyOnPartialReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::getcrosskvcachefraction (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig23getCrossKvCacheFractionEv", false]], "tensorrt_llm::executor::kvcacheconfig::getenableblockreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig19getEnableBlockReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::getenablepartialreuse (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEnablePartialReuseEv", false]], "tensorrt_llm::executor::kvcacheconfig::geteventbuffermaxsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEventBufferMaxSizeEv", false]], "tensorrt_llm::executor::kvcacheconfig::getfreegpumemoryfraction (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getFreeGpuMemoryFractionEv", false]], "tensorrt_llm::executor::kvcacheconfig::gethostcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getHostCacheSizeEv", false]], "tensorrt_llm::executor::kvcacheconfig::getmaxattentionwindowvec (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getMaxAttentionWindowVecEv", false]], "tensorrt_llm::executor::kvcacheconfig::getmaxtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig12getMaxTokensEv", false]], "tensorrt_llm::executor::kvcacheconfig::getonboardblocks (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getOnboardBlocksEv", false]], "tensorrt_llm::executor::kvcacheconfig::getsecondaryoffloadminpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig30getSecondaryOffloadMinPriorityEv", false]], "tensorrt_llm::executor::kvcacheconfig::getsinktokenlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig18getSinkTokenLengthEv", false]], "tensorrt_llm::executor::kvcacheconfig::kvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", false]], "tensorrt_llm::executor::kvcacheconfig::mcopyonpartialreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mCopyOnPartialReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::mcrosskvcachefraction (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21mCrossKvCacheFractionE", false]], "tensorrt_llm::executor::kvcacheconfig::menableblockreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig17mEnableBlockReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::menablepartialreuse (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEnablePartialReuseE", false]], "tensorrt_llm::executor::kvcacheconfig::meventbuffermaxsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEventBufferMaxSizeE", false]], "tensorrt_llm::executor::kvcacheconfig::mfreegpumemoryfraction (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mFreeGpuMemoryFractionE", false]], "tensorrt_llm::executor::kvcacheconfig::mhostcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mHostCacheSizeE", false]], "tensorrt_llm::executor::kvcacheconfig::mmaxattentionwindowvec (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mMaxAttentionWindowVecE", false]], "tensorrt_llm::executor::kvcacheconfig::mmaxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig10mMaxTokensE", false]], "tensorrt_llm::executor::kvcacheconfig::monboardblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mOnboardBlocksE", false]], "tensorrt_llm::executor::kvcacheconfig::msecondaryoffloadminpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig28mSecondaryOffloadMinPriorityE", false]], "tensorrt_llm::executor::kvcacheconfig::msinktokenlength (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16mSinkTokenLengthE", false]], "tensorrt_llm::executor::kvcacheconfig::setcopyonpartialreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::setcrosskvcachefraction (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", false]], "tensorrt_llm::executor::kvcacheconfig::setenableblockreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::setenablepartialreuse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", false]], "tensorrt_llm::executor::kvcacheconfig::seteventbuffermaxsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", false]], "tensorrt_llm::executor::kvcacheconfig::setfreegpumemoryfraction (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", false]], "tensorrt_llm::executor::kvcacheconfig::sethostcachesize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", false]], "tensorrt_llm::executor::kvcacheconfig::setmaxattentionwindowvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::kvcacheconfig::setmaxtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", false]], "tensorrt_llm::executor::kvcacheconfig::setonboardblocks (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", false]], "tensorrt_llm::executor::kvcacheconfig::setsecondaryoffloadminpriority (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", false]], "tensorrt_llm::executor::kvcacheconfig::setsinktokenlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", false]], "tensorrt_llm::executor::kvcachecreateddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedDataE", false]], "tensorrt_llm::executor::kvcachecreateddata::numblockspercachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedData22numBlocksPerCacheLevelE", false]], "tensorrt_llm::executor::kvcacheevent (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEventE", false]], "tensorrt_llm::executor::kvcacheevent::data (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent4dataE", false]], "tensorrt_llm::executor::kvcacheevent::eventid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent7eventIdE", false]], "tensorrt_llm::executor::kvcacheevent::kvcacheevent (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", false]], "tensorrt_llm::executor::kvcacheeventdata (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDataE", false]], "tensorrt_llm::executor::kvcacheeventdiff (c++ struct)": [[0, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", false]], "tensorrt_llm::executor::kvcacheeventdiff::newvalue (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8newValueE", false]], "tensorrt_llm::executor::kvcacheeventdiff::oldvalue (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8oldValueE", false]], "tensorrt_llm::executor::kvcacheeventmanager (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManagerE", false]], "tensorrt_llm::executor::kvcacheeventmanager::getlatestevents (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::kvcacheeventmanager::kvcacheeventmanager (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", false]], "tensorrt_llm::executor::kvcacheeventmanager::kvcachemanager (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager14kvCacheManagerE", false]], "tensorrt_llm::executor::kvcacheremoveddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedDataE", false]], "tensorrt_llm::executor::kvcacheremoveddata::blockhashes (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedData11blockHashesE", false]], "tensorrt_llm::executor::kvcacheretentionconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfigE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getdecodedurationms (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig19getDecodeDurationMsEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getdecoderetentionpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig26getDecodeRetentionPriorityEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getdirectory (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig12getDirectoryEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::getperblockretentionpriorityduration (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcacheretentionconfig::gettokenrangeretentionconfigs (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig29getTokenRangeRetentionConfigsEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::gettransfermode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig15getTransferModeEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kdefaultretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25kDefaultRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kmaxretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMaxRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kminretentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMinRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::kvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", false], [0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigEv", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mdecodedurationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig17mDecodeDurationMsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mdecoderetentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig24mDecodeRetentionPriorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mdirectory (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig10mDirectoryE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mtokenrangeretentionconfigs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig27mTokenRangeRetentionConfigsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::mtransfermode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig13mTransferModeE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::durationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10durationMsE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8priorityE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8tokenEndE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenrangeretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::kvcacheretentionconfig::tokenrangeretentionconfig::tokenstart (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10tokenStartE", false]], "tensorrt_llm::executor::kvcachestats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStatsE", false]], "tensorrt_llm::executor::kvcachestats::allocnewblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14allocNewBlocksE", false]], "tensorrt_llm::executor::kvcachestats::alloctotalblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats16allocTotalBlocksE", false]], "tensorrt_llm::executor::kvcachestats::cachehitrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12cacheHitRateE", false]], "tensorrt_llm::executor::kvcachestats::freenumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13freeNumBlocksE", false]], "tensorrt_llm::executor::kvcachestats::maxnumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12maxNumBlocksE", false]], "tensorrt_llm::executor::kvcachestats::missedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12missedBlocksE", false]], "tensorrt_llm::executor::kvcachestats::reusedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12reusedBlocksE", false]], "tensorrt_llm::executor::kvcachestats::tokensperblock (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14tokensPerBlockE", false]], "tensorrt_llm::executor::kvcachestats::usednumblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13usedNumBlocksE", false]], "tensorrt_llm::executor::kvcachestoredblockdata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockDataE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::blockhash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData9blockHashE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::cachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData10cacheLevelE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::kvcachestoredblockdata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcachestoredblockdata::loraid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6loraIdE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData8priorityE", false]], "tensorrt_llm::executor::kvcachestoredblockdata::tokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6tokensE", false]], "tensorrt_llm::executor::kvcachestoreddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredDataE", false]], "tensorrt_llm::executor::kvcachestoreddata::blocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData6blocksE", false]], "tensorrt_llm::executor::kvcachestoreddata::parenthash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData10parentHashE", false]], "tensorrt_llm::executor::kvcachetransfermode (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferModeE", false]], "tensorrt_llm::executor::kvcachetransfermode::dram (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode4DRAME", false]], "tensorrt_llm::executor::kvcachetransfermode::gds (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode3GDSE", false]], "tensorrt_llm::executor::kvcachetransfermode::posix_debug_fallback (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode20POSIX_DEBUG_FALLBACKE", false]], "tensorrt_llm::executor::kvcacheupdateddata (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedDataE", false]], "tensorrt_llm::executor::kvcacheupdateddata::blockhash (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData9blockHashE", false]], "tensorrt_llm::executor::kvcacheupdateddata::cachelevel (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData10cacheLevelE", false]], "tensorrt_llm::executor::kvcacheupdateddata::cachelevelupdated (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::kvcacheupdateddata::kvcacheupdateddata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", false]], "tensorrt_llm::executor::kvcacheupdateddata::priority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData8priorityE", false]], "tensorrt_llm::executor::kvcacheupdateddata::priorityupdated (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", false]], "tensorrt_llm::executor::logitspostprocessor (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor19LogitsPostProcessorE", false]], "tensorrt_llm::executor::logitspostprocessorbatched (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor26LogitsPostProcessorBatchedE", false]], "tensorrt_llm::executor::logitspostprocessorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfigE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getprocessorbatched (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig19getProcessorBatchedEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getprocessormap (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig15getProcessorMapEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::getreplicate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig12getReplicateEv", false]], "tensorrt_llm::executor::logitspostprocessorconfig::logitspostprocessorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mprocessorbatched (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig17mProcessorBatchedE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mprocessormap (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig13mProcessorMapE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::mreplicate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig10mReplicateE", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setprocessorbatched (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setprocessormap (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", false]], "tensorrt_llm::executor::logitspostprocessorconfig::setreplicate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", false]], "tensorrt_llm::executor::logitspostprocessormap (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor22LogitsPostProcessorMapE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfigE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::calculatespeculativeresource (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig28calculateSpeculativeResourceEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::calculatespeculativeresourcetuple (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::get (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig3getEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig12getNgramSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getverificationsetsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig22getVerificationSetSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::getwindowsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig13getWindowSizeEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::isle (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::islegal (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingngram (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig30kDefaultLookaheadDecodingNgramE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingverificationset (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig40kDefaultLookaheadDecodingVerificationSetE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::kdefaultlookaheaddecodingwindow (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig31kDefaultLookaheadDecodingWindowE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::lookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", false], [0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigEv", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig10mNgramSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mverificationsetsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig20mVerificationSetSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::mwindowsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig11mWindowSizeE", false]], "tensorrt_llm::executor::lookaheaddecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::loraconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfigE", false]], "tensorrt_llm::executor::loraconfig::getconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getConfigEv", false]], "tensorrt_llm::executor::loraconfig::gettaskid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getTaskIdEv", false]], "tensorrt_llm::executor::loraconfig::getweights (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor10LoraConfig10getWeightsEv", false]], "tensorrt_llm::executor::loraconfig::loraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", false]], "tensorrt_llm::executor::loraconfig::mconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mConfigE", false]], "tensorrt_llm::executor::loraconfig::mtaskid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mTaskIdE", false]], "tensorrt_llm::executor::loraconfig::mweights (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10LoraConfig8mWeightsE", false]], "tensorrt_llm::executor::medusachoices (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor13MedusaChoicesE", false]], "tensorrt_llm::executor::memorytype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryTypeE", false]], "tensorrt_llm::executor::memorytype::kcpu (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kCPUE", false]], "tensorrt_llm::executor::memorytype::kcpu_pinned (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType11kCPU_PINNEDE", false]], "tensorrt_llm::executor::memorytype::kcpu_pinnedpool (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType15kCPU_PINNEDPOOLE", false]], "tensorrt_llm::executor::memorytype::kgpu (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kGPUE", false]], "tensorrt_llm::executor::memorytype::kunknown (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType8kUNKNOWNE", false]], "tensorrt_llm::executor::memorytype::kuvm (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor10MemoryType4kUVME", false]], "tensorrt_llm::executor::millisecondstype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16MillisecondsTypeE", false]], "tensorrt_llm::executor::modeltype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelTypeE", false]], "tensorrt_llm::executor::modeltype::kdecoder_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType13kDECODER_ONLYE", false]], "tensorrt_llm::executor::modeltype::kencoder_decoder (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType16kENCODER_DECODERE", false]], "tensorrt_llm::executor::modeltype::kencoder_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor9ModelType13kENCODER_ONLYE", false]], "tensorrt_llm::executor::mropeconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfigE", false]], "tensorrt_llm::executor::mropeconfig::getmropepositiondeltas (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11MropeConfig22getMRopePositionDeltasEv", false]], "tensorrt_llm::executor::mropeconfig::getmroperotarycossin (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor11MropeConfig20getMRopeRotaryCosSinEv", false]], "tensorrt_llm::executor::mropeconfig::mmropepositiondeltas (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig20mMRopePositionDeltasE", false]], "tensorrt_llm::executor::mropeconfig::mmroperotarycossin (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig18mMRopeRotaryCosSinE", false]], "tensorrt_llm::executor::mropeconfig::mropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", false]], "tensorrt_llm::executor::operator<< (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", false], [0, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", false]], "tensorrt_llm::executor::orchestratorconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfigE", false]], "tensorrt_llm::executor::orchestratorconfig::getisorchestrator (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getIsOrchestratorEv", false]], "tensorrt_llm::executor::orchestratorconfig::getorchleadercomm (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getOrchLeaderCommEv", false]], "tensorrt_llm::executor::orchestratorconfig::getspawnprocesses (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getSpawnProcessesEv", false]], "tensorrt_llm::executor::orchestratorconfig::getworkerexecutablepath (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig23getWorkerExecutablePathEv", false]], "tensorrt_llm::executor::orchestratorconfig::misorchestrator (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mIsOrchestratorE", false]], "tensorrt_llm::executor::orchestratorconfig::morchleadercomm (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mOrchLeaderCommE", false]], "tensorrt_llm::executor::orchestratorconfig::mspawnprocesses (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mSpawnProcessesE", false]], "tensorrt_llm::executor::orchestratorconfig::mworkerexecutablepath (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig21mWorkerExecutablePathE", false]], "tensorrt_llm::executor::orchestratorconfig::orchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", false]], "tensorrt_llm::executor::orchestratorconfig::setisorchestrator (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", false]], "tensorrt_llm::executor::orchestratorconfig::setorchleadercomm (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", false]], "tensorrt_llm::executor::orchestratorconfig::setspawnprocesses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", false]], "tensorrt_llm::executor::orchestratorconfig::setworkerexecutablepath (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", false]], "tensorrt_llm::executor::outputconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfigE", false]], "tensorrt_llm::executor::outputconfig::additionalmodeloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22additionalModelOutputsE", false]], "tensorrt_llm::executor::outputconfig::excludeinputfromoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22excludeInputFromOutputE", false]], "tensorrt_llm::executor::outputconfig::outputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", false]], "tensorrt_llm::executor::outputconfig::returncontextlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnContextLogitsE", false]], "tensorrt_llm::executor::outputconfig::returnencoderoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnEncoderOutputE", false]], "tensorrt_llm::executor::outputconfig::returngenerationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig22returnGenerationLogitsE", false]], "tensorrt_llm::executor::outputconfig::returnlogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig14returnLogProbsE", false]], "tensorrt_llm::executor::outputconfig::returnperfmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12OutputConfig17returnPerfMetricsE", false]], "tensorrt_llm::executor::parallelconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfigE", false]], "tensorrt_llm::executor::parallelconfig::getcommunicationmode (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationModeEv", false]], "tensorrt_llm::executor::parallelconfig::getcommunicationtype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationTypeEv", false]], "tensorrt_llm::executor::parallelconfig::getdeviceids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig12getDeviceIdsEv", false]], "tensorrt_llm::executor::parallelconfig::getnumnodes (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig11getNumNodesEv", false]], "tensorrt_llm::executor::parallelconfig::getorchestratorconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig21getOrchestratorConfigEv", false]], "tensorrt_llm::executor::parallelconfig::getparticipantids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig17getParticipantIdsEv", false]], "tensorrt_llm::executor::parallelconfig::mcommmode (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommModeE", false]], "tensorrt_llm::executor::parallelconfig::mcommtype (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommTypeE", false]], "tensorrt_llm::executor::parallelconfig::mdeviceids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig10mDeviceIdsE", false]], "tensorrt_llm::executor::parallelconfig::mnumnodes (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mNumNodesE", false]], "tensorrt_llm::executor::parallelconfig::morchestratorconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig19mOrchestratorConfigE", false]], "tensorrt_llm::executor::parallelconfig::mparticipantids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig15mParticipantIdsE", false]], "tensorrt_llm::executor::parallelconfig::parallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::parallelconfig::setcommunicationmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", false]], "tensorrt_llm::executor::parallelconfig::setcommunicationtype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", false]], "tensorrt_llm::executor::parallelconfig::setdeviceids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::parallelconfig::setnumnodes (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", false]], "tensorrt_llm::executor::parallelconfig::setorchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", false]], "tensorrt_llm::executor::parallelconfig::setparticipantids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::peftcacheconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfigE", false]], "tensorrt_llm::executor::peftcacheconfig::getdevicecachepercent (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getDeviceCachePercentEv", false]], "tensorrt_llm::executor::peftcacheconfig::gethostcachesize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getHostCacheSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::getloraprefetchdir (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig18getLoraPrefetchDirEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxadaptersize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getMaxAdapterSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxpagesperblockdevice (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig25getMaxPagesPerBlockDeviceEv", false]], "tensorrt_llm::executor::peftcacheconfig::getmaxpagesperblockhost (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getMaxPagesPerBlockHostEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumcopystreams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getNumCopyStreamsEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumdevicemodulelayer (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getNumDeviceModuleLayerEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumensureworkers (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig19getNumEnsureWorkersEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumhostmodulelayer (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getNumHostModuleLayerEv", false]], "tensorrt_llm::executor::peftcacheconfig::getnumputworkers (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getNumPutWorkersEv", false]], "tensorrt_llm::executor::peftcacheconfig::getoptimaladaptersize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getOptimalAdapterSizeEv", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxadaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig22kDefaultMaxAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxpagesperblockdevice (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig30kDefaultMaxPagesPerBlockDeviceE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultmaxpagesperblockhost (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig28kDefaultMaxPagesPerBlockHostE", false]], "tensorrt_llm::executor::peftcacheconfig::kdefaultoptimaladaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig26kDefaultOptimalAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mdevicecachepercent (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mDeviceCachePercentE", false]], "tensorrt_llm::executor::peftcacheconfig::mhostcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mHostCacheSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mloraprefetchdir (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig16mLoraPrefetchDirE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxadaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mMaxAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxpagesperblockdevice (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig23mMaxPagesPerBlockDeviceE", false]], "tensorrt_llm::executor::peftcacheconfig::mmaxpagesperblockhost (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mMaxPagesPerBlockHostE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumcopystreams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mNumCopyStreamsE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumdevicemodulelayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mNumDeviceModuleLayerE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumensureworkers (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig17mNumEnsureWorkersE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumhostmodulelayer (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mNumHostModuleLayerE", false]], "tensorrt_llm::executor::peftcacheconfig::mnumputworkers (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mNumPutWorkersE", false]], "tensorrt_llm::executor::peftcacheconfig::moptimaladaptersize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mOptimalAdapterSizeE", false]], "tensorrt_llm::executor::peftcacheconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", false]], "tensorrt_llm::executor::peftcacheconfig::peftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", false]], "tensorrt_llm::executor::prioritytype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor12PriorityTypeE", false]], "tensorrt_llm::executor::prompttuningconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfigE", false]], "tensorrt_llm::executor::prompttuningconfig::getembeddingtable (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig17getEmbeddingTableEv", false]], "tensorrt_llm::executor::prompttuningconfig::getinputtokenextraids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig21getInputTokenExtraIdsEv", false]], "tensorrt_llm::executor::prompttuningconfig::membeddingtable (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig15mEmbeddingTableE", false]], "tensorrt_llm::executor::prompttuningconfig::minputtokenextraids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig19mInputTokenExtraIdsE", false]], "tensorrt_llm::executor::prompttuningconfig::prompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", false]], "tensorrt_llm::executor::randomseedtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor14RandomSeedTypeE", false]], "tensorrt_llm::executor::request (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestE", false]], "tensorrt_llm::executor::request::getadditionaloutputnames (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request24getAdditionalOutputNamesEv", false]], "tensorrt_llm::executor::request::getallottedtimems (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request17getAllottedTimeMsEv", false]], "tensorrt_llm::executor::request::getbadwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getBadWordsEv", false]], "tensorrt_llm::executor::request::getclientid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getClientIdEv", false]], "tensorrt_llm::executor::request::getcontextphaseparams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getContextPhaseParamsEv", false]], "tensorrt_llm::executor::request::getcrossattentionmask (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getCrossAttentionMaskEv", false]], "tensorrt_llm::executor::request::geteagleconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getEagleConfigEv", false]], "tensorrt_llm::executor::request::getembeddingbias (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request16getEmbeddingBiasEv", false]], "tensorrt_llm::executor::request::getencoderinputfeatures (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputFeaturesEv", false]], "tensorrt_llm::executor::request::getencoderinputtokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputTokenIdsEv", false]], "tensorrt_llm::executor::request::getencoderoutputlength (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getEncoderOutputLengthEv", false]], "tensorrt_llm::executor::request::getendid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request8getEndIdEv", false]], "tensorrt_llm::executor::request::getexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request28getExternalDraftTokensConfigEv", false]], "tensorrt_llm::executor::request::getguideddecodingparams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request23getGuidedDecodingParamsEv", false]], "tensorrt_llm::executor::request::getinputtokenids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request16getInputTokenIdsEv", false]], "tensorrt_llm::executor::request::getkvcacheretentionconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request25getKvCacheRetentionConfigEv", false]], "tensorrt_llm::executor::request::getlanguageadapteruid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getLanguageAdapterUidEv", false]], "tensorrt_llm::executor::request::getlogitspostprocessor (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getLogitsPostProcessorEv", false]], "tensorrt_llm::executor::request::getlogitspostprocessorname (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request26getLogitsPostProcessorNameEv", false]], "tensorrt_llm::executor::request::getlookaheadconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request18getLookaheadConfigEv", false]], "tensorrt_llm::executor::request::getloraconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request13getLoraConfigEv", false]], "tensorrt_llm::executor::request::getmaxtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getMaxTokensEv", false]], "tensorrt_llm::executor::request::getmropeconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getMropeConfigEv", false]], "tensorrt_llm::executor::request::getmultimodalembedding (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getMultimodalEmbeddingEv", false]], "tensorrt_llm::executor::request::getoutputconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request15getOutputConfigEv", false]], "tensorrt_llm::executor::request::getpadid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request8getPadIdEv", false]], "tensorrt_llm::executor::request::getpositionids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getPositionIdsEv", false]], "tensorrt_llm::executor::request::getpriority (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request11getPriorityEv", false]], "tensorrt_llm::executor::request::getprompttuningconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request21getPromptTuningConfigEv", false]], "tensorrt_llm::executor::request::getrequesttype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request14getRequestTypeEv", false]], "tensorrt_llm::executor::request::getreturnallgeneratedtokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request27getReturnAllGeneratedTokensEv", false]], "tensorrt_llm::executor::request::getsamplingconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request17getSamplingConfigEv", false]], "tensorrt_llm::executor::request::getskipcrossattnblocks (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request22getSkipCrossAttnBlocksEv", false]], "tensorrt_llm::executor::request::getstopwords (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getStopWordsEv", false]], "tensorrt_llm::executor::request::getstreaming (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor7Request12getStreamingEv", false]], "tensorrt_llm::executor::request::kbatchedpostprocessorname (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request25kBatchedPostProcessorNameE", false]], "tensorrt_llm::executor::request::kdefaultpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request16kDefaultPriorityE", false]], "tensorrt_llm::executor::request::kdynamicpostprocessornameprefix (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request31kDynamicPostProcessorNamePrefixE", false]], "tensorrt_llm::executor::request::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor7Request5mImplE", false]], "tensorrt_llm::executor::request::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", false]], "tensorrt_llm::executor::request::request (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", false]], "tensorrt_llm::executor::request::setallottedtimems (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", false]], "tensorrt_llm::executor::request::setbadwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", false]], "tensorrt_llm::executor::request::setclientid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", false]], "tensorrt_llm::executor::request::setcontextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", false]], "tensorrt_llm::executor::request::setcrossattentionmask (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", false]], "tensorrt_llm::executor::request::seteagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", false]], "tensorrt_llm::executor::request::setembeddingbias (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", false]], "tensorrt_llm::executor::request::setencoderinputfeatures (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", false]], "tensorrt_llm::executor::request::setencoderinputtokenids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", false]], "tensorrt_llm::executor::request::setencoderoutputlength (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", false]], "tensorrt_llm::executor::request::setendid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", false]], "tensorrt_llm::executor::request::setexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", false]], "tensorrt_llm::executor::request::setguideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", false]], "tensorrt_llm::executor::request::setkvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", false]], "tensorrt_llm::executor::request::setlanguageadapteruid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", false]], "tensorrt_llm::executor::request::setlogitspostprocessor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", false]], "tensorrt_llm::executor::request::setlogitspostprocessorname (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", false]], "tensorrt_llm::executor::request::setlookaheadconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", false]], "tensorrt_llm::executor::request::setloraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", false]], "tensorrt_llm::executor::request::setmropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", false]], "tensorrt_llm::executor::request::setmultimodalembedding (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", false]], "tensorrt_llm::executor::request::setoutputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", false]], "tensorrt_llm::executor::request::setpadid (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", false]], "tensorrt_llm::executor::request::setpositionids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::executor::request::setpriority (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", false]], "tensorrt_llm::executor::request::setprompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", false]], "tensorrt_llm::executor::request::setrequesttype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", false]], "tensorrt_llm::executor::request::setreturnallgeneratedtokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", false]], "tensorrt_llm::executor::request::setsamplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", false]], "tensorrt_llm::executor::request::setskipcrossattnblocks (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", false]], "tensorrt_llm::executor::request::setstopwords (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", false]], "tensorrt_llm::executor::request::setstreaming (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", false]], "tensorrt_llm::executor::request::~request (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7RequestD0Ev", false]], "tensorrt_llm::executor::requestperfmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::firstiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9firstIterE", false]], "tensorrt_llm::executor::requestperfmetrics::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics4iterE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14kvCacheMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::kvcachehitrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics14kvCacheHitRateE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::nummissedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numMissedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numnewallocatedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics21numNewAllocatedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numreusedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numReusedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::kvcachemetrics::numtotalallocatedblocks (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics23numTotalAllocatedBlocksE", false]], "tensorrt_llm::executor::requestperfmetrics::lastiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics8lastIterE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecoding (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics19speculativeDecodingE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::acceptancerate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics14acceptanceRateE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::totalaccepteddrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics24totalAcceptedDraftTokensE", false]], "tensorrt_llm::executor::requestperfmetrics::speculativedecodingmetrics::totaldrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics16totalDraftTokensE", false]], "tensorrt_llm::executor::requestperfmetrics::timepoint (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9TimePointE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13timingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetricsE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::arrivaltime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11arrivalTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::firstscheduledtime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18firstScheduledTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::firsttokentime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics14firstTokenTimeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachesize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11kvCacheSizeE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachetransferend (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18kvCacheTransferEndE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::kvcachetransferstart (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics20kvCacheTransferStartE", false]], "tensorrt_llm::executor::requestperfmetrics::timingmetrics::lasttokentime (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics13lastTokenTimeE", false]], "tensorrt_llm::executor::requeststage (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStageE", false]], "tensorrt_llm::executor::requeststage::kcontext_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kCONTEXT_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kencoder_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kENCODER_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kgeneration_complete (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage20kGENERATION_COMPLETEE", false]], "tensorrt_llm::executor::requeststage::kgeneration_in_progress (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage23kGENERATION_IN_PROGRESSE", false]], "tensorrt_llm::executor::requeststage::kqueued (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStage7kQUEUEDE", false]], "tensorrt_llm::executor::requeststats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStatsE", false]], "tensorrt_llm::executor::requeststats::allocnewblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats24allocNewBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::alloctotalblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats26allocTotalBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::avgnumdecodedtokensperiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats26avgNumDecodedTokensPerIterE", false]], "tensorrt_llm::executor::requeststats::contextprefillposition (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22contextPrefillPositionE", false]], "tensorrt_llm::executor::requeststats::disservingstats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats15disServingStatsE", false]], "tensorrt_llm::executor::requeststats::id (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats2idE", false]], "tensorrt_llm::executor::requeststats::kvcachehitrateperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats24kvCacheHitRatePerRequestE", false]], "tensorrt_llm::executor::requeststats::missedblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22missedBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::numgeneratedtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats18numGeneratedTokensE", false]], "tensorrt_llm::executor::requeststats::paused (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats6pausedE", false]], "tensorrt_llm::executor::requeststats::reusedblocksperrequest (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats22reusedBlocksPerRequestE", false]], "tensorrt_llm::executor::requeststats::scheduled (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats9scheduledE", false]], "tensorrt_llm::executor::requeststats::stage (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor12RequestStats5stageE", false]], "tensorrt_llm::executor::requeststatsperiteration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIterationE", false]], "tensorrt_llm::executor::requeststatsperiteration::iter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration4iterE", false]], "tensorrt_llm::executor::requeststatsperiteration::requeststats (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration12requestStatsE", false]], "tensorrt_llm::executor::requesttype (c++ enum)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestTypeE", false]], "tensorrt_llm::executor::requesttype::request_type_context_and_generation (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType35REQUEST_TYPE_CONTEXT_AND_GENERATIONE", false]], "tensorrt_llm::executor::requesttype::request_type_context_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType25REQUEST_TYPE_CONTEXT_ONLYE", false]], "tensorrt_llm::executor::requesttype::request_type_generation_only (c++ enumerator)": [[0, "_CPPv4N12tensorrt_llm8executor11RequestType28REQUEST_TYPE_GENERATION_ONLYE", false]], "tensorrt_llm::executor::response (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseE", false]], "tensorrt_llm::executor::response::getclientid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response11getClientIdEv", false]], "tensorrt_llm::executor::response::geterrormsg (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response11getErrorMsgEv", false]], "tensorrt_llm::executor::response::getrequestid (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response12getRequestIdEv", false]], "tensorrt_llm::executor::response::getresult (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response9getResultEv", false]], "tensorrt_llm::executor::response::haserror (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor8Response8hasErrorEv", false]], "tensorrt_llm::executor::response::mimpl (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor8Response5mImplE", false]], "tensorrt_llm::executor::response::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", false]], "tensorrt_llm::executor::response::response (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", false]], "tensorrt_llm::executor::response::~response (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor8ResponseD0Ev", false]], "tensorrt_llm::executor::result (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor6ResultE", false]], "tensorrt_llm::executor::result::additionaloutputs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result17additionalOutputsE", false]], "tensorrt_llm::executor::result::contextlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13contextLogitsE", false]], "tensorrt_llm::executor::result::contextphaseparams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result18contextPhaseParamsE", false]], "tensorrt_llm::executor::result::cumlogprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result11cumLogProbsE", false]], "tensorrt_llm::executor::result::decodingiter (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result12decodingIterE", false]], "tensorrt_llm::executor::result::encoderoutput (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13encoderOutputE", false]], "tensorrt_llm::executor::result::finishreasons (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13finishReasonsE", false]], "tensorrt_llm::executor::result::generationlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result16generationLogitsE", false]], "tensorrt_llm::executor::result::isfinal (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result7isFinalE", false]], "tensorrt_llm::executor::result::issequencefinal (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result15isSequenceFinalE", false]], "tensorrt_llm::executor::result::logprobs (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result8logProbsE", false]], "tensorrt_llm::executor::result::outputtokenids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result14outputTokenIdsE", false]], "tensorrt_llm::executor::result::requestperfmetrics (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result18requestPerfMetricsE", false]], "tensorrt_llm::executor::result::sequenceindex (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result13sequenceIndexE", false]], "tensorrt_llm::executor::result::specdecfastlogitsinfo (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Result21specDecFastLogitsInfoE", false]], "tensorrt_llm::executor::retentionpriority (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor17RetentionPriorityE", false]], "tensorrt_llm::executor::retentionpriorityandduration (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDurationE", false]], "tensorrt_llm::executor::retentionpriorityandduration::durationms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration10durationMsE", false]], "tensorrt_llm::executor::retentionpriorityandduration::retentionpriority (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration17retentionPriorityE", false]], "tensorrt_llm::executor::retentionpriorityandduration::retentionpriorityandduration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", false]], "tensorrt_llm::executor::samplingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfigE", false]], "tensorrt_llm::executor::samplingconfig::checkbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkbeamwidtharray (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkearlystopping (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checklengthpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checkmintokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checknorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::checknumreturnsequences (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::checkrepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktopk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppdecay (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppmin (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::checktoppresetids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", false]], "tensorrt_llm::executor::samplingconfig::getbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig26getBeamSearchDiversityRateEv", false]], "tensorrt_llm::executor::samplingconfig::getbeamwidth (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getBeamWidthEv", false]], "tensorrt_llm::executor::samplingconfig::getbeamwidtharray (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getBeamWidthArrayEv", false]], "tensorrt_llm::executor::samplingconfig::getearlystopping (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getEarlyStoppingEv", false]], "tensorrt_llm::executor::samplingconfig::getfrequencypenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig19getFrequencyPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getlengthpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getLengthPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getminp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getMinPEv", false]], "tensorrt_llm::executor::samplingconfig::getmintokens (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getMinTokensEv", false]], "tensorrt_llm::executor::samplingconfig::getnorepeatngramsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getNoRepeatNgramSizeEv", false]], "tensorrt_llm::executor::samplingconfig::getnumreturnbeams (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getNumReturnBeamsEv", false]], "tensorrt_llm::executor::samplingconfig::getnumreturnsequences (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig21getNumReturnSequencesEv", false]], "tensorrt_llm::executor::samplingconfig::getpresencepenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig18getPresencePenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getrepetitionpenalty (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getRepetitionPenaltyEv", false]], "tensorrt_llm::executor::samplingconfig::getseed (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getSeedEv", false]], "tensorrt_llm::executor::samplingconfig::gettemperature (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig14getTemperatureEv", false]], "tensorrt_llm::executor::samplingconfig::gettopk (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopKEv", false]], "tensorrt_llm::executor::samplingconfig::gettopp (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopPEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppdecay (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getTopPDecayEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppmin (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig10getTopPMinEv", false]], "tensorrt_llm::executor::samplingconfig::gettoppresetids (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig15getTopPResetIdsEv", false]], "tensorrt_llm::executor::samplingconfig::mbeamsearchdiversityrate (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig24mBeamSearchDiversityRateE", false]], "tensorrt_llm::executor::samplingconfig::mbeamwidth (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mBeamWidthE", false]], "tensorrt_llm::executor::samplingconfig::mbeamwidtharray (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mBeamWidthArrayE", false]], "tensorrt_llm::executor::samplingconfig::mearlystopping (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mEarlyStoppingE", false]], "tensorrt_llm::executor::samplingconfig::mfrequencypenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17mFrequencyPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mlengthpenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mLengthPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mminp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mMinPE", false]], "tensorrt_llm::executor::samplingconfig::mmintokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mMinTokensE", false]], "tensorrt_llm::executor::samplingconfig::mnorepeatngramsize (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mNoRepeatNgramSizeE", false]], "tensorrt_llm::executor::samplingconfig::mnumreturnbeams (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mNumReturnBeamsE", false]], "tensorrt_llm::executor::samplingconfig::mnumreturnsequences (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19mNumReturnSequencesE", false]], "tensorrt_llm::executor::samplingconfig::mpresencepenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16mPresencePenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mrepetitionpenalty (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mRepetitionPenaltyE", false]], "tensorrt_llm::executor::samplingconfig::mseed (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mSeedE", false]], "tensorrt_llm::executor::samplingconfig::mtemperature (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12mTemperatureE", false]], "tensorrt_llm::executor::samplingconfig::mtopk (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopKE", false]], "tensorrt_llm::executor::samplingconfig::mtopp (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopPE", false]], "tensorrt_llm::executor::samplingconfig::mtoppdecay (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mTopPDecayE", false]], "tensorrt_llm::executor::samplingconfig::mtoppmin (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig8mTopPMinE", false]], "tensorrt_llm::executor::samplingconfig::mtoppresetids (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig13mTopPResetIdsE", false]], "tensorrt_llm::executor::samplingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", false]], "tensorrt_llm::executor::samplingconfig::samplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", false]], "tensorrt_llm::executor::samplingconfig::setbeamsearchdiversityrate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setbeamwidth (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", false]], "tensorrt_llm::executor::samplingconfig::setbeamwidtharray (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", false]], "tensorrt_llm::executor::samplingconfig::setearlystopping (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setfrequencypenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setlengthpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setminp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setmintokens (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setnorepeatngramsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setnumreturnsequences (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::setpresencepenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setrepetitionpenalty (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::setseed (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settemperature (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settopk (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", false]], "tensorrt_llm::executor::samplingconfig::settopp (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppdecay (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppmin (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", false]], "tensorrt_llm::executor::samplingconfig::settoppresetids (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", false]], "tensorrt_llm::executor::samplingconfig::updatenumreturnbeams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20updateNumReturnBeamsEv", false]], "tensorrt_llm::executor::schedulerconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfigE", false]], "tensorrt_llm::executor::schedulerconfig::getcapacityschedulerpolicy (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig26getCapacitySchedulerPolicyEv", false]], "tensorrt_llm::executor::schedulerconfig::getcontextchunkingpolicy (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig24getContextChunkingPolicyEv", false]], "tensorrt_llm::executor::schedulerconfig::getdynamicbatchconfig (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig21getDynamicBatchConfigEv", false]], "tensorrt_llm::executor::schedulerconfig::mcapacityschedulerpolicy (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig24mCapacitySchedulerPolicyE", false]], "tensorrt_llm::executor::schedulerconfig::mcontextchunkingpolicy (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig22mContextChunkingPolicyE", false]], "tensorrt_llm::executor::schedulerconfig::mdynamicbatchconfig (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig19mDynamicBatchConfigE", false]], "tensorrt_llm::executor::schedulerconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", false]], "tensorrt_llm::executor::schedulerconfig::schedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", false]], "tensorrt_llm::executor::serialization (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor13SerializationE", false]], "tensorrt_llm::executor::serialization::deserializeadditionalmodeloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeadditionaloutput (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeagentstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeAgentStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializebool (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecachestate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecachetransceiverconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecommstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializecontextphaseparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedatatransceiverstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedebugconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedecodingmode (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedisservingrequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializedynamicbatchconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeeagleconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeexecutorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeextendedruntimeperfknobconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeexternaldrafttokensconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeguideddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeguideddecodingparams (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeinflightbatchingstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeiterationstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeiterationstatsvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializekvcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializekvcacheretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializekvcachestats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializelookaheaddecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeloraconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializemodeltype (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializemropeconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeorchestratorconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeoutputconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeparallelconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializepeftcacheconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeprompttuningconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequest (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequestperfmetrics (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststage (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststatsperiteration (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializerequeststatsperiterationvec (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializeresponse (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeresponses (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", false]], "tensorrt_llm::executor::serialization::deserializeresult (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializesamplingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializeschedulerconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializesocketstate (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializespecdecfastlogitsinfo (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializespecdecodingstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeSpecDecodingStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializespeculativedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializestaticbatchingstats (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializestring (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetimepoint (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::deserializetokenrangeretentionconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", false]], "tensorrt_llm::executor::serialization::serialize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK17SpecDecodingStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10AgentStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", false]], "tensorrt_llm::executor::serialization::serializedsize (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK17SpecDecodingStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN18RequestPerfMetrics9TimePointE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10AgentStateE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", false], [0, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", false]], "tensorrt_llm::executor::shape (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor5ShapeE", false]], "tensorrt_llm::executor::shape::base (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape4BaseE", false]], "tensorrt_llm::executor::shape::dimtype64 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape9DimType64E", false]], "tensorrt_llm::executor::shape::shape (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", false], [0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", false], [0, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEv", false]], "tensorrt_llm::executor::sizetype32 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10SizeType32E", false]], "tensorrt_llm::executor::sizetype64 (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor10SizeType64E", false]], "tensorrt_llm::executor::specdecodingstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStatsE", false]], "tensorrt_llm::executor::specdecodingstats::acceptancelength (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats16acceptanceLengthE", false]], "tensorrt_llm::executor::specdecodingstats::draftoverhead (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats13draftOverheadE", false]], "tensorrt_llm::executor::specdecodingstats::iterlatencyms (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats13iterLatencyMSE", false]], "tensorrt_llm::executor::specdecodingstats::numacceptedtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats17numAcceptedTokensE", false]], "tensorrt_llm::executor::specdecodingstats::numdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats14numDraftTokensE", false]], "tensorrt_llm::executor::specdecodingstats::numrequestswithdrafttokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats26numRequestsWithDraftTokensE", false]], "tensorrt_llm::executor::speculativedecodingconfig (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfigE", false]], "tensorrt_llm::executor::speculativedecodingconfig::fastlogits (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig10fastLogitsE", false]], "tensorrt_llm::executor::speculativedecodingconfig::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", false]], "tensorrt_llm::executor::speculativedecodingconfig::speculativedecodingconfig (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfoE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::draftparticipantid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo18draftParticipantIdE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::draftrequestid (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo14draftRequestIdE", false]], "tensorrt_llm::executor::speculativedecodingfastlogitsinfo::totensor (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo8toTensorEv", false]], "tensorrt_llm::executor::staticbatchingstats (c++ struct)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStatsE", false]], "tensorrt_llm::executor::staticbatchingstats::emptygenslots (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats13emptyGenSlotsE", false]], "tensorrt_llm::executor::staticbatchingstats::numcontextrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats18numContextRequestsE", false]], "tensorrt_llm::executor::staticbatchingstats::numctxtokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numCtxTokensE", false]], "tensorrt_llm::executor::staticbatchingstats::numgentokens (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numGenTokensE", false]], "tensorrt_llm::executor::staticbatchingstats::numscheduledrequests (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats20numScheduledRequestsE", false]], "tensorrt_llm::executor::streamptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9StreamPtrE", false]], "tensorrt_llm::executor::tensor (c++ class)": [[0, "_CPPv4N12tensorrt_llm8executor6TensorE", false]], "tensorrt_llm::executor::tensor::copyto (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::copytocpu (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytogpu (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytomanaged (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytopinned (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::copytopooledpinned (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::cpu (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::cudastreamptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor13CudaStreamPtrE", false]], "tensorrt_llm::executor::tensor::detail::ofitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", false]], "tensorrt_llm::executor::tensor::detail::toitensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", false]], "tensorrt_llm::executor::tensor::getdata (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7getDataEv", false], [0, "_CPPv4NK12tensorrt_llm8executor6Tensor7getDataEv", false]], "tensorrt_llm::executor::tensor::getdatatype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor11getDataTypeEv", false]], "tensorrt_llm::executor::tensor::getmemorytype (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor13getMemoryTypeEv", false]], "tensorrt_llm::executor::tensor::getruntimetype (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", false]], "tensorrt_llm::executor::tensor::getshape (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor8getShapeEv", false]], "tensorrt_llm::executor::tensor::getsize (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor7getSizeEv", false]], "tensorrt_llm::executor::tensor::getsizeinbytes (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6Tensor14getSizeInBytesEv", false]], "tensorrt_llm::executor::tensor::gpu (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", false]], "tensorrt_llm::executor::tensor::impl (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor4ImplE", false]], "tensorrt_llm::executor::tensor::managed (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::mtensor (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7mTensorE", false]], "tensorrt_llm::executor::tensor::of (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", false], [0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", false]], "tensorrt_llm::executor::tensor::operator bool (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensorcvbEv", false]], "tensorrt_llm::executor::tensor::operator!= (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", false]], "tensorrt_llm::executor::tensor::operator= (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", false]], "tensorrt_llm::executor::tensor::operator== (c++ function)": [[0, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", false]], "tensorrt_llm::executor::tensor::pinned (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::pooledpinned (c++ function)": [[0, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", false]], "tensorrt_llm::executor::tensor::setfrom (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::setzero (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", false]], "tensorrt_llm::executor::tensor::tensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", false], [0, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorEv", false]], "tensorrt_llm::executor::tensor::~tensor (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor6TensorD0Ev", false]], "tensorrt_llm::executor::tensorptr (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9TensorPtrE", false]], "tensorrt_llm::executor::tokenidtype (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11TokenIdTypeE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIbEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIbE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIfEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIfE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsI4halfEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsI4halfE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int32_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int32_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int64_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int64_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt6int8_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt6int8_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEE5valueE", false]], "tensorrt_llm::executor::typetraits (c++ struct)": [[0, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", false]], "tensorrt_llm::executor::typetraits::value (c++ member)": [[0, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIP1TE5valueE", false]], "tensorrt_llm::executor::veclogprobs (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor11VecLogProbsE", false]], "tensorrt_llm::executor::vectokenextraids (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor16VecTokenExtraIdsE", false]], "tensorrt_llm::executor::vectokens (c++ type)": [[0, "_CPPv4N12tensorrt_llm8executor9VecTokensE", false]], "tensorrt_llm::executor::version (c++ function)": [[0, "_CPPv4N12tensorrt_llm8executor7versionEv", false]], "tensorrt_llm::layers (c++ type)": [[1, "_CPPv4N12tensorrt_llm6layersE", false]], "tensorrt_llm::mpi (c++ type)": [[0, "_CPPv4N12tensorrt_llm3mpiE", false]], "tensorrt_llm::runtime (c++ type)": [[0, "_CPPv4N12tensorrt_llm7runtimeE", false], [1, "_CPPv4N12tensorrt_llm7runtimeE", false]], "tensorrt_llm::runtime::allreducebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffersE", false]], "tensorrt_llm::runtime::allreducebuffers::allreducebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", false]], "tensorrt_llm::runtime::allreducebuffers::mallreducecommptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers18mAllReduceCommPtrsE", false]], "tensorrt_llm::runtime::allreducebuffers::mflagptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9mFlagPtrsE", false]], "tensorrt_llm::runtime::allreducebuffers::mipcmemoryhandles (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers17mIpcMemoryHandlesE", false]], "tensorrt_llm::runtime::allreducebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::buffercast (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", false]], "tensorrt_llm::runtime::buffercastornull (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", false]], "tensorrt_llm::runtime::bufferdatatype (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::bufferdatatype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", false]], "tensorrt_llm::runtime::bufferdatatype::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType11getDataTypeEv", false]], "tensorrt_llm::runtime::bufferdatatype::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType7getSizeEv", false]], "tensorrt_llm::runtime::bufferdatatype::getsizeinbits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType13getSizeInBitsEv", false]], "tensorrt_llm::runtime::bufferdatatype::ispointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType9isPointerEv", false]], "tensorrt_llm::runtime::bufferdatatype::isunsigned (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType10isUnsignedEv", false]], "tensorrt_llm::runtime::bufferdatatype::ktrtpointertype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType15kTrtPointerTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mDataTypeE", false]], "tensorrt_llm::runtime::bufferdatatype::mpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType8mPointerE", false]], "tensorrt_llm::runtime::bufferdatatype::munsigned (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mUnsignedE", false]], "tensorrt_llm::runtime::bufferdatatype::operator nvinfer1::datatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataTypecvN8nvinfer18DataTypeEEv", false]], "tensorrt_llm::runtime::buffermanager (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerE", false]], "tensorrt_llm::runtime::buffermanager::allocate (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::buffermanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", false]], "tensorrt_llm::runtime::buffermanager::copy (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", false]], "tensorrt_llm::runtime::buffermanager::copyfrom (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", false]], "tensorrt_llm::runtime::buffermanager::cpu (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::cudamempoolptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager14CudaMemPoolPtrE", false]], "tensorrt_llm::runtime::buffermanager::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13CudaStreamPtrE", false]], "tensorrt_llm::runtime::buffermanager::emptybuffer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::emptytensor (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::getstream (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager9getStreamEv", false]], "tensorrt_llm::runtime::buffermanager::gpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::gpusync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::ibufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10IBufferPtrE", false]], "tensorrt_llm::runtime::buffermanager::ipcnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::itensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10ITensorPtrE", false]], "tensorrt_llm::runtime::buffermanager::kbyte_type (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10kBYTE_TYPEE", false]], "tensorrt_llm::runtime::buffermanager::managed (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::memorypoolfree (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolFreeEv", false]], "tensorrt_llm::runtime::buffermanager::memorypoolreserved (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager18memoryPoolReservedEv", false]], "tensorrt_llm::runtime::buffermanager::memorypooltrimto (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", false]], "tensorrt_llm::runtime::buffermanager::memorypoolused (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolUsedEv", false]], "tensorrt_llm::runtime::buffermanager::mpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager5mPoolE", false]], "tensorrt_llm::runtime::buffermanager::mstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7mStreamE", false]], "tensorrt_llm::runtime::buffermanager::mtrimpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager9mTrimPoolE", false]], "tensorrt_llm::runtime::buffermanager::pinned (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::pinnedpool (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", false], [1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::buffermanager::setmem (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", false]], "tensorrt_llm::runtime::buffermanager::setzero (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", false]], "tensorrt_llm::runtime::buffermanager::~buffermanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerD0Ev", false]], "tensorrt_llm::runtime::bufferrange (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", false]], "tensorrt_llm::runtime::bufferrange::base (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11BufferRange4BaseE", false]], "tensorrt_llm::runtime::bufferrange::bufferrange (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", false], [1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", false]], "tensorrt_llm::runtime::canaccesspeer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", false]], "tensorrt_llm::runtime::constpointercast (c++ function)": [[1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", false]], "tensorrt_llm::runtime::cudaevent (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEventE", false]], "tensorrt_llm::runtime::cudaevent::cudaevent (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", false], [1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", false]], "tensorrt_llm::runtime::cudaevent::deleter (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7DeleterE", false]], "tensorrt_llm::runtime::cudaevent::deleter::deleter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", false], [1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEv", false]], "tensorrt_llm::runtime::cudaevent::deleter::mownsevent (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter10mOwnsEventE", false]], "tensorrt_llm::runtime::cudaevent::deleter::operator() (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", false]], "tensorrt_llm::runtime::cudaevent::element_type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent12element_typeE", false]], "tensorrt_llm::runtime::cudaevent::eventptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent8EventPtrE", false]], "tensorrt_llm::runtime::cudaevent::get (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent3getEv", false]], "tensorrt_llm::runtime::cudaevent::mevent (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent6mEventE", false]], "tensorrt_llm::runtime::cudaevent::pointer (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7pointerE", false]], "tensorrt_llm::runtime::cudaevent::synchronize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent11synchronizeEv", false]], "tensorrt_llm::runtime::cudastream (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStreamE", false]], "tensorrt_llm::runtime::cudastream::cudastream (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", false]], "tensorrt_llm::runtime::cudastream::deleter (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7DeleterE", false]], "tensorrt_llm::runtime::cudastream::deleter::deleter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", false], [1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEv", false]], "tensorrt_llm::runtime::cudastream::deleter::mownsstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter11mOwnsStreamE", false]], "tensorrt_llm::runtime::cudastream::deleter::operator() (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", false]], "tensorrt_llm::runtime::cudastream::get (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream3getEv", false]], "tensorrt_llm::runtime::cudastream::getdevice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream9getDeviceEv", false]], "tensorrt_llm::runtime::cudastream::mdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mDeviceE", false]], "tensorrt_llm::runtime::cudastream::mstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mStreamE", false]], "tensorrt_llm::runtime::cudastream::record (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", false]], "tensorrt_llm::runtime::cudastream::streamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10CudaStream9StreamPtrE", false]], "tensorrt_llm::runtime::cudastream::synchronize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream11synchronizeEv", false]], "tensorrt_llm::runtime::cudastream::wait (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4typeE", false]], "tensorrt_llm::runtime::datatypetraits (c++ struct)": [[1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", false]], "tensorrt_llm::runtime::datatypetraits::name (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4nameE", false]], "tensorrt_llm::runtime::datatypetraits::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4sizeE", false]], "tensorrt_llm::runtime::datatypetraits::type (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4typeE", false]], "tensorrt_llm::runtime::decoder (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoderE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffersE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::beamsearchbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::mcumlogprobstmp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers15mCumLogProbsTmpE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::mnumsms (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7mNumSMsE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::moutputbeamhypotheses (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers21mOutputBeamHypothesesE", false]], "tensorrt_llm::runtime::decoder::beamsearchbuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decoder::decoderstate (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderStateE", false]], "tensorrt_llm::runtime::decoder::decoderstate::allocatespeculativedecodingbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::decoderstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::decodinginputptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16DecodingInputPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::decodingoutputptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState17DecodingOutputPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", false]], "tensorrt_llm::runtime::decoder::decoderstate::getacceptedlengthscumsum (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24getAcceptedLengthsCumSumEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getacceptedpackedpaths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getAcceptedPackedPathsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getallnewtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getAllNewTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getbeamsearchbuffers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getBeamSearchBuffersEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getcumlogprobs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishedsteps (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishedStepsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishedsum (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getFinishedSumEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getfinishreasons (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishReasonsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getgatheredids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getjointdecodinginput (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState21getJointDecodingInputEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getjointdecodingoutput (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getJointDecodingOutputEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getlogprobs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxbatchsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBatchSizeEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxdecodingdecodertokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState27getMaxDecodingDecoderTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxdecodingenginetokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getMaxDecodingEngineTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getmaxsequencelength (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getMaxSequenceLengthEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnextdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getNextDraftTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnextdrafttokenslengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getNextDraftTokensLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getnumdecodingenginetokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getparentids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState12getParentIdsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getprevdrafttokenslengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getPrevDraftTokensLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getsequencelengths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsE10SizeType32", false], [1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::getspeculativedecodingmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getSpeculativeDecodingModeEv", false]], "tensorrt_llm::runtime::decoder::decoderstate::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13LlmRequestPtrE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mbeamsearchbuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mBeamSearchBuffersE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mfinishedsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState14mFinishedStepsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mjointdecodinginput (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState19mJointDecodingInputE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mjointdecodingoutput (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState20mJointDecodingOutputE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxbeamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBeamWidthE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxdecodingdecodertokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState25mMaxDecodingDecoderTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxdecodingenginetokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mMaxDecodingEngineTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mmaxsequencelength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mMaxSequenceLengthE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mnumdecodingenginetokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mNumDecodingEngineTokensE", false]], "tensorrt_llm::runtime::decoder::decoderstate::mspeculativedecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mSpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::decoder::decoderstate::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13RequestVectorE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setnumdecodingenginetokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decoder::decoderstate::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupeagle (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupexplicitdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", false]], "tensorrt_llm::runtime::decoder::decoderstate::setuplookahead (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", false]], "tensorrt_llm::runtime::decoder::decoderstate::setupspeculativedecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::decoder::decoderstate::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", false]], "tensorrt_llm::runtime::decoder_batch::input (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5InputE", false]], "tensorrt_llm::runtime::decoder_batch::input::batchslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input10batchSlotsE", false]], "tensorrt_llm::runtime::decoder_batch::input::batchslotsrequestorder (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input22batchSlotsRequestOrderE", false]], "tensorrt_llm::runtime::decoder_batch::input::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input16cacheIndirectionE", false]], "tensorrt_llm::runtime::decoder_batch::input::eagleinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input11eagleInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::eaglelastinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15eagleLastInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::explicitdrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input25explicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::explicitdrafttokenslastinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input29explicitDraftTokensLastInputsE", false]], "tensorrt_llm::runtime::decoder_batch::input::generationsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15generationStepsE", false]], "tensorrt_llm::runtime::decoder_batch::input::input (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", false], [1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", false]], "tensorrt_llm::runtime::decoder_batch::input::logits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input6logitsE", false]], "tensorrt_llm::runtime::decoder_batch::input::maxdecodersteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15maxDecoderStepsE", false]], "tensorrt_llm::runtime::decoder_batch::input::predicteddraftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input20predictedDraftLogitsE", false]], "tensorrt_llm::runtime::decoder_batch::input::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input14TensorConstPtrE", false]], "tensorrt_llm::runtime::decoder_batch::input::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch::output (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6OutputE", false]], "tensorrt_llm::runtime::decoder_batch::output::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output16cacheIndirectionE", false]], "tensorrt_llm::runtime::decoder_batch::output::output (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output6OutputEv", false]], "tensorrt_llm::runtime::decoder_batch::output::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output9TensorPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7RequestE", false]], "tensorrt_llm::runtime::decoder_batch::request::badwordslist (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12badWordsListE", false]], "tensorrt_llm::runtime::decoder_batch::request::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9BufferPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request::draftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftLogitsE", false]], "tensorrt_llm::runtime::decoder_batch::request::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftTokensE", false]], "tensorrt_llm::runtime::decoder_batch::request::dtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5dtypeE", false]], "tensorrt_llm::runtime::decoder_batch::request::eagleconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11eagleConfigE", false]], "tensorrt_llm::runtime::decoder_batch::request::embeddingbias (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13embeddingBiasE", false]], "tensorrt_llm::runtime::decoder_batch::request::endid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5endIdE", false]], "tensorrt_llm::runtime::decoder_batch::request::generatedtokensperenginestep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request28generatedTokensPerEngineStepE", false]], "tensorrt_llm::runtime::decoder_batch::request::ids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request3idsE", false]], "tensorrt_llm::runtime::decoder_batch::request::inputlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request8inputLenE", false]], "tensorrt_llm::runtime::decoder_batch::request::lookaheadruntimeconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request22lookaheadRuntimeConfigE", false]], "tensorrt_llm::runtime::decoder_batch::request::maxnewtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12maxNewTokensE", false]], "tensorrt_llm::runtime::decoder_batch::request::medusapaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11medusaPathsE", false]], "tensorrt_llm::runtime::decoder_batch::request::medusatreeids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13medusaTreeIdsE", false]], "tensorrt_llm::runtime::decoder_batch::request::request (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::decoder_batch::request::stopwordslist (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13stopWordsListE", false]], "tensorrt_llm::runtime::decoder_batch::request::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request14TensorConstPtrE", false]], "tensorrt_llm::runtime::decoder_batch::request::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9TensorPtrE", false]], "tensorrt_llm::runtime::decodinginput (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInputE", false]], "tensorrt_llm::runtime::decodinginput::badwordslens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsLensE", false]], "tensorrt_llm::runtime::decodinginput::badwordslists (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13badWordsListsE", false]], "tensorrt_llm::runtime::decodinginput::badwordsptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsPtrsE", false]], "tensorrt_llm::runtime::decodinginput::batchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9batchSizeE", false]], "tensorrt_llm::runtime::decodinginput::batchslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10batchSlotsE", false]], "tensorrt_llm::runtime::decodinginput::beamwidths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10beamWidthsE", false]], "tensorrt_llm::runtime::decodinginput::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput16cacheIndirectionE", false]], "tensorrt_llm::runtime::decodinginput::decodinginput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11eagleInputsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs12acceptedLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedpathids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15acceptedPathIdsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::acceptedtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14acceptedTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::eagleinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13lastDraftLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14lastDraftPathsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::lastdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15lastDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13nextDraftLensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14nextDraftPathsE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::eagleinputs::seqslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs8seqSlotsE", false]], "tensorrt_llm::runtime::decodinginput::embeddingbias (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13embeddingBiasE", false]], "tensorrt_llm::runtime::decodinginput::endids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6endIdsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25explicitDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::bestpathindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::bestpathlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16lastDraftIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15lastDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21lastGenerationLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::lastpositionidsbase (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs19lastPositionIdsBaseE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::masks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs5masksE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::maxgenlengthdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs18maxGenLengthDeviceE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16nextDraftIndicesE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdraftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextDraftProbsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextflattokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextFlatTokensE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::nextgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21nextGenerationLengthsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::packedpositionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs17packedPositionIdsE", false]], "tensorrt_llm::runtime::decodinginput::explicitdrafttokensinputs::seqslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs8seqSlotsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25externalDraftTokensInputsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::constantthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs17constantThresholdE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::draftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11draftLogitsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::draftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs10draftProbsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::drafttokenids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs13draftTokenIdsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::numdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14numDraftTokensE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::numdrafttokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18numDraftTokensHostE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::step (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs4stepE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::targetprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11targetProbsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::usedraftlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14useDraftLogitsE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::usedraftlogitshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18useDraftLogitsHostE", false]], "tensorrt_llm::runtime::decodinginput::externaldrafttokensinputs::userandomacceptancethreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs28useRandomAcceptanceThresholdE", false]], "tensorrt_llm::runtime::decodinginput::finishreasons (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13finishReasonsE", false]], "tensorrt_llm::runtime::decodinginput::generationsteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15generationStepsE", false]], "tensorrt_llm::runtime::decodinginput::lengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput7lengthsE", false]], "tensorrt_llm::runtime::decodinginput::logits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6logitsE", false]], "tensorrt_llm::runtime::decodinginput::logitsvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9logitsVecE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15lookaheadInputsE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputsE", false]], "tensorrt_llm::runtime::decodinginput::lookaheadinputs::tokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputs13tokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::maxattentionwindow (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput18maxAttentionWindowE", false]], "tensorrt_llm::runtime::decodinginput::maxbadwordslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14maxBadWordsLenE", false]], "tensorrt_llm::runtime::decodinginput::maxlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9maxLengthE", false]], "tensorrt_llm::runtime::decodinginput::maxstopwordslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15maxStopWordsLenE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12medusaInputsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusacurtokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs22medusaCurTokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusalogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs12medusaLogitsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusapaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs11medusaPathsE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusatargettokensperstep (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs25medusaTargetTokensPerStepE", false]], "tensorrt_llm::runtime::decodinginput::medusainputs::medusatreeids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs13medusaTreeIdsE", false]], "tensorrt_llm::runtime::decodinginput::norepeatngramsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput17noRepeatNgramSizeE", false]], "tensorrt_llm::runtime::decodinginput::sequencelimitlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput19sequenceLimitLengthE", false]], "tensorrt_llm::runtime::decodinginput::sinktokenlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15sinkTokenLengthE", false]], "tensorrt_llm::runtime::decodinginput::step (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput4stepE", false]], "tensorrt_llm::runtime::decodinginput::stopwordslens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsLensE", false]], "tensorrt_llm::runtime::decodinginput::stopwordslists (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14stopWordsListsE", false]], "tensorrt_llm::runtime::decodinginput::stopwordsptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsPtrsE", false]], "tensorrt_llm::runtime::decodinginput::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14TensorConstPtrE", false]], "tensorrt_llm::runtime::decodinginput::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9TensorPtrE", false]], "tensorrt_llm::runtime::decodingoutput (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutputE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypothesesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14beamHypothesesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::batchdones (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses10batchDonesE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::cumlogprobscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses14cumLogProbsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::empty (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::init (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::logprobscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11logProbsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::minnormedscorescba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18minNormedScoresCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::normedscorescba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses15normedScoresCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::numbeamscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11numBeamsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::outputidscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses12outputIdsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::release (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7releaseEv", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::sequencelengthscba (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18sequenceLengthsCBAE", false]], "tensorrt_llm::runtime::decodingoutput::beamhypotheses::slice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::decodingoutput::cacheindirection (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16cacheIndirectionE", false]], "tensorrt_llm::runtime::decodingoutput::cumlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11cumLogProbsE", false]], "tensorrt_llm::runtime::decodingoutput::decodingoutput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::decodingoutput::eaglebuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12eagleBuffersE", false]], "tensorrt_llm::runtime::decodingoutput::explicitdrafttokensbuffers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26explicitDraftTokensBuffersE", false]], "tensorrt_llm::runtime::decodingoutput::finishedsum (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11finishedSumE", false]], "tensorrt_llm::runtime::decodingoutput::finishreasons (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13finishReasonsE", false]], "tensorrt_llm::runtime::decodingoutput::gatheredids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11gatheredIdsE", false]], "tensorrt_llm::runtime::decodingoutput::ids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput3idsE", false]], "tensorrt_llm::runtime::decodingoutput::knegativeinfinity (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput17kNegativeInfinityE", false]], "tensorrt_llm::runtime::decodingoutput::lengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput7lengthsE", false]], "tensorrt_llm::runtime::decodingoutput::logprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput8logProbsE", false]], "tensorrt_llm::runtime::decodingoutput::logprobstiled (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13logProbsTiledE", false]], "tensorrt_llm::runtime::decodingoutput::lookaheadoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16lookaheadOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::newtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9newTokensE", false]], "tensorrt_llm::runtime::decodingoutput::newtokenssteps (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14newTokensStepsE", false]], "tensorrt_llm::runtime::decodingoutput::newtokensvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12newTokensVecE", false]], "tensorrt_llm::runtime::decodingoutput::parentids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9parentIdsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26speculativeDecodingOutputsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::acceptedlengthscumsum (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs21acceptedLengthsCumSumE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::acceptedtokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs17acceptedTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::nextdrafttokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18nextDraftTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::pathsoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs12pathsOffsetsE", false]], "tensorrt_llm::runtime::decodingoutput::speculativedecodingoutputs::prevdrafttokenslen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18prevDraftTokensLenE", false]], "tensorrt_llm::runtime::decodingoutput::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9TensorPtrE", false]], "tensorrt_llm::runtime::deviceallocationnvls (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", false]], "tensorrt_llm::runtime::deviceallocationnvls::_capacity (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls9_capacityE", false]], "tensorrt_llm::runtime::deviceallocationnvls::_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls7_handleE", false]], "tensorrt_llm::runtime::deviceallocationnvls::deviceallocationnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls20DeviceAllocationNvlsEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::free (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls4freeEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getcapacity (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls11getCapacityEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getipcunicastpointers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls21getIpcUnicastPointersEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getmulticastpointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls19getMulticastPointerEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::getunicastpointer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls17getUnicastPointerEv", false]], "tensorrt_llm::runtime::deviceallocationnvls::reset (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", false]], "tensorrt_llm::runtime::deviceallocationnvls::~deviceallocationnvls (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvlsD0Ev", false]], "tensorrt_llm::runtime::eaglebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffersE", false]], "tensorrt_llm::runtime::eaglebuffers::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9BufferPtrE", false]], "tensorrt_llm::runtime::eaglebuffers::chunkedcontextnexttokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers28chunkedContextNextTokensHostE", false]], "tensorrt_llm::runtime::eaglebuffers::cumsumgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers23cumSumGenerationLengthsE", false]], "tensorrt_llm::runtime::eaglebuffers::eaglebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::engineinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12engineInputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13engineOutputsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs12acceptedLensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13acceptedPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::acceptedtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14acceptedTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdraftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13nextDraftLensE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdraftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14nextDraftPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::engineoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::greedysamplinghost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18greedySamplingHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6InputsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersdrafttokenids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22allLayersDraftTokenIdsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersdrafttokenidspredecessor (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33allLayersDraftTokenIdsPredecessorE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::alllayersscores (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15allLayersScoresE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::chunkedcontextnexttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs24chunkedContextNextTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::currentexpandindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20currentExpandIndicesE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftlens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs9draftLensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10draftPathsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::draftpathshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14draftPathsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs11draftTokensE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::dynamictreemaxtopkhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22dynamicTreeMaxTopKHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxcontextlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetCtxContextLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxpastkeyvaluelengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetCtxPastKeyValueLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetctxrequesttypeshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetCtxRequestTypesHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgencontextlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetGenContextLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgenpastkeyvaluelengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetGenPastKeyValueLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::eaglenetgenrequesttypeshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetGenRequestTypesHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::inputgentokenshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18inputGenTokensHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::posterioralpha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14posteriorAlphaE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::posteriorthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18posteriorThresholdE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::prevscores (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10prevScoresE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::randomdatasample (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs16randomDataSampleE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::randomdatavalidation (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20randomDataValidationE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodinggenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29specDecodingGenerationLengthsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodinggenerationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33specDecodingGenerationLengthsHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodingpackedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs23specDecodingPackedMasksE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::specdecodingpositionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27specDecodingPositionOffsetsE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::temperatures (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs12temperaturesE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::usedynamictreehost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18useDynamicTreeHostE", false]], "tensorrt_llm::runtime::eaglebuffers::inputs::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15useSpecDecodingE", false]], "tensorrt_llm::runtime::eaglebuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::itensor (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7ITensorE", false]], "tensorrt_llm::runtime::eaglebuffers::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13LlmRequestPtrE", false]], "tensorrt_llm::runtime::eaglebuffers::maxgenerationlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers19maxGenerationLengthE", false]], "tensorrt_llm::runtime::eaglebuffers::mdefaultposteriorthreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26mDefaultPosteriorThresholdE", false]], "tensorrt_llm::runtime::eaglebuffers::mdogreedysampling (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers17mDoGreedySamplingE", false]], "tensorrt_llm::runtime::eaglebuffers::posterioralphahost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18posteriorAlphaHostE", false]], "tensorrt_llm::runtime::eaglebuffers::posteriorthresholdhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers22posteriorThresholdHostE", false]], "tensorrt_llm::runtime::eaglebuffers::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13RequestVectorE", false]], "tensorrt_llm::runtime::eaglebuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::scanreducetempstorage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers21scanReduceTempStorageE", false]], "tensorrt_llm::runtime::eaglebuffers::scanreducetempstoragebytes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26scanReduceTempStorageBytesE", false]], "tensorrt_llm::runtime::eaglebuffers::setfrominputs (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", false], [1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::eaglebuffers::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers10SizeType32E", false]], "tensorrt_llm::runtime::eaglebuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorMapE", false]], "tensorrt_llm::runtime::eaglebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::eaglemodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModuleE", false]], "tensorrt_llm::runtime::eaglemodule::eaglemodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleEv", false]], "tensorrt_llm::runtime::eaglemodule::getdefaulteaglechoices (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule22getDefaultEagleChoicesEv", false]], "tensorrt_llm::runtime::eaglemodule::getmaxnonleafnodesperlayer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule26getMaxNonLeafNodesPerLayerEv", false]], "tensorrt_llm::runtime::eaglemodule::getnumtransformerlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule23getNumTransformerLayersEv", false]], "tensorrt_llm::runtime::eaglemodule::mdefaulteaglechoices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule20mDefaultEagleChoicesE", false]], "tensorrt_llm::runtime::eaglemodule::mmaxnonleafnodesperlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule24mMaxNonLeafNodesPerLayerE", false]], "tensorrt_llm::runtime::eaglemodule::mnumtransformerslayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11EagleModule21mNumTransformersLayerE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffersE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9BufferPtrE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::cumsumgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers23cumSumGenerationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12engineInputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs::positionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs15positionOffsetsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineinputs::requesttypesdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs18requestTypesDeviceE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13engineOutputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::bestpathindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::bestpathlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::masks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs5masksE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::maxgentoken (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs11maxGenTokenE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdraftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs16nextDraftIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdraftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextDraftProbsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15nextDraftTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextflattokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextFlatTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextgenerationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs21nextGenerationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::nextpositionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs19nextPositionOffsetsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::packedpositionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs17packedPositionIdsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::engineoutputs::totalgentoken (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs13totalGenTokenE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::explicitdrafttokensbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6InputsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::draftindices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12draftIndicesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::draftprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs10draftProbsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::drafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11draftTokensE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::generationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs17generationLengthsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::generationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs21generationLengthsHostE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::maxgenlengthhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16maxGenLengthHostE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::packedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11packedMasksE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::positionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11positionIdsE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::positionidsbase (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15positionIdsBaseE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::randomdatasample (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16randomDataSampleE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::randomdatavalidation (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs20randomDataValidationE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::temperatures (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12temperaturesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::inputs::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15useSpecDecodingE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::itensor (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7ITensorE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::scantempstorage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers15scanTempStorageE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::scantempstoragebytes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers20scanTempStorageBytesE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::setfrominputs (c++ function)": [[1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", false], [1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers10SizeType32E", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorMapE", false]], "tensorrt_llm::runtime::explicitdrafttokensbuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::genericprompttuningparams (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", false]], "tensorrt_llm::runtime::genericprompttuningparams::embeddingtable (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams14embeddingTableE", false]], "tensorrt_llm::runtime::genericprompttuningparams::genericprompttuningparams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::genericprompttuningparams::prompttuningenabled (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams19promptTuningEnabledE", false]], "tensorrt_llm::runtime::genericprompttuningparams::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams10SizeType32E", false]], "tensorrt_llm::runtime::genericprompttuningparams::tasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams5tasksE", false]], "tensorrt_llm::runtime::genericprompttuningparams::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9TensorPtrE", false]], "tensorrt_llm::runtime::genericprompttuningparams::vocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9vocabSizeE", false]], "tensorrt_llm::runtime::getdefaultbatchslots (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", false]], "tensorrt_llm::runtime::gptdecoder (c++ class)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", false]], "tensorrt_llm::runtime::gptdecoder::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13CudaStreamPtrE", false]], "tensorrt_llm::runtime::gptdecoder::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", false]], "tensorrt_llm::runtime::gptdecoder::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::gptdecoder::forwardsync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::gptdecoder::getsamplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder17getSamplingConfigEv", false]], "tensorrt_llm::runtime::gptdecoder::gptdecoder (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::gptdecoder::mdecodinglayerworkspace (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder23mDecodingLayerWorkspaceE", false]], "tensorrt_llm::runtime::gptdecoder::mdecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mDecodingModeE", false]], "tensorrt_llm::runtime::gptdecoder::mdynamicdecodelayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder19mDynamicDecodeLayerE", false]], "tensorrt_llm::runtime::gptdecoder::mmanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder8mManagerE", false]], "tensorrt_llm::runtime::gptdecoder::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::gptdecoder::msamplingconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder15mSamplingConfigE", false]], "tensorrt_llm::runtime::gptdecoder::mvocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10mVocabSizeE", false]], "tensorrt_llm::runtime::gptdecoder::mvocabsizepadded (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16mVocabSizePaddedE", false]], "tensorrt_llm::runtime::gptdecoder::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", false]], "tensorrt_llm::runtime::gptdecoder::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder9TensorPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatchedE", false]], "tensorrt_llm::runtime::gptdecoderbatched::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13CudaStreamPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", false]], "tensorrt_llm::runtime::gptdecoderbatched::finalize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", false]], "tensorrt_llm::runtime::gptdecoderbatched::forward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::forwarddispatch (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::getbuffermanager (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getBufferManagerEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getdecoderstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getdecoderstream (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getDecoderStreamEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::getunderlyingdecoder (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched20getUnderlyingDecoderEv", false]], "tensorrt_llm::runtime::gptdecoderbatched::gptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::gptdecoderbatched::gptdecoderptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13GptDecoderPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13LlmRequestPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mbuffermanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mBufferManagerE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoder (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched8mDecoderE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoderstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13mDecoderStateE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mdecoderstream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mDecoderStreamE", false]], "tensorrt_llm::runtime::gptdecoderbatched::mruntimestream (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mRuntimeStreamE", false]], "tensorrt_llm::runtime::gptdecoderbatched::prepareforward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13RequestVectorE", false]], "tensorrt_llm::runtime::gptdecoderbatched::seteagleinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::setexplicitdrafttokensinputs (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::gptdecoderbatched::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::gptdecoderbatched::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14SharedConstPtrE", false]], "tensorrt_llm::runtime::gptdecoderbatched::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched9TensorPtrE", false]], "tensorrt_llm::runtime::gptjsonconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfigE", false]], "tensorrt_llm::runtime::gptjsonconfig::enginefilename (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", false], [1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", false]], "tensorrt_llm::runtime::gptjsonconfig::getcontextparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig21getContextParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getgpuspernode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getGpusPerNodeEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getmodelconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getModelConfigEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getmodelconfigmutable (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig21getModelConfigMutableEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getname (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig7getNameEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getpipelineparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig22getPipelineParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getprecision (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getPrecisionEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getruntimedefaults (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig18getRuntimeDefaultsEv", false]], "tensorrt_llm::runtime::gptjsonconfig::gettensorparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig20getTensorParallelismEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getversion (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig10getVersionEv", false]], "tensorrt_llm::runtime::gptjsonconfig::getworldsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getWorldSizeEv", false]], "tensorrt_llm::runtime::gptjsonconfig::gptjsonconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", false]], "tensorrt_llm::runtime::gptjsonconfig::mcontextparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig19mContextParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mGpusPerNodeE", false]], "tensorrt_llm::runtime::gptjsonconfig::mmodelconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mModelConfigE", false]], "tensorrt_llm::runtime::gptjsonconfig::mname (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5mNameE", false]], "tensorrt_llm::runtime::gptjsonconfig::mpipelineparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig20mPipelineParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mprecision (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig10mPrecisionE", false]], "tensorrt_llm::runtime::gptjsonconfig::mruntimedefaults (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig16mRuntimeDefaultsE", false]], "tensorrt_llm::runtime::gptjsonconfig::mtensorparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig18mTensorParallelismE", false]], "tensorrt_llm::runtime::gptjsonconfig::mversion (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig8mVersionE", false]], "tensorrt_llm::runtime::gptjsonconfig::parse (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", false], [1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", false], [1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", false]], "tensorrt_llm::runtime::ibuffer (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferE", false]], "tensorrt_llm::runtime::ibuffer::data (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataEv", false]], "tensorrt_llm::runtime::ibuffer::datatype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer8DataTypeE", false]], "tensorrt_llm::runtime::ibuffer::getcapacity (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getCapacityEv", false]], "tensorrt_llm::runtime::ibuffer::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getDataTypeEv", false]], "tensorrt_llm::runtime::ibuffer::getdatatypename (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", false], [1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer15getDataTypeNameEv", false]], "tensorrt_llm::runtime::ibuffer::getmemorytype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer13getMemoryTypeEv", false]], "tensorrt_llm::runtime::ibuffer::getmemorytypename (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer17getMemoryTypeNameEv", false]], "tensorrt_llm::runtime::ibuffer::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7getSizeEv", false]], "tensorrt_llm::runtime::ibuffer::getsizeinbytes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer14getSizeInBytesEv", false]], "tensorrt_llm::runtime::ibuffer::ibuffer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferERK7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferEv", false]], "tensorrt_llm::runtime::ibuffer::memorytype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", false]], "tensorrt_llm::runtime::ibuffer::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferaSERK7IBuffer", false]], "tensorrt_llm::runtime::ibuffer::release (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7releaseEv", false]], "tensorrt_llm::runtime::ibuffer::resize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14SharedConstPtrE", false]], "tensorrt_llm::runtime::ibuffer::sharedptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9SharedPtrE", false]], "tensorrt_llm::runtime::ibuffer::slice (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::tobytes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::uniqueconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14UniqueConstPtrE", false]], "tensorrt_llm::runtime::ibuffer::uniqueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9UniquePtrE", false]], "tensorrt_llm::runtime::ibuffer::view (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::wrap (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", false]], "tensorrt_llm::runtime::ibuffer::~ibuffer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7IBufferD0Ev", false]], "tensorrt_llm::runtime::igptdecoder (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderE", false]], "tensorrt_llm::runtime::igptdecoder::create (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::igptdecoder::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", false]], "tensorrt_llm::runtime::igptdecoder::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::igptdecoder::forwardsync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", false]], "tensorrt_llm::runtime::igptdecoder::getsamplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder17getSamplingConfigEv", false]], "tensorrt_llm::runtime::igptdecoder::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", false]], "tensorrt_llm::runtime::igptdecoder::tensorconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder14TensorConstPtrE", false]], "tensorrt_llm::runtime::igptdecoder::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder9TensorPtrE", false]], "tensorrt_llm::runtime::igptdecoder::~igptdecoder (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderD0Ev", false]], "tensorrt_llm::runtime::igptdecoderbatched (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedE", false]], "tensorrt_llm::runtime::igptdecoderbatched::cudastreamptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13CudaStreamPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::disablelookahead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", false]], "tensorrt_llm::runtime::igptdecoderbatched::finalize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", false]], "tensorrt_llm::runtime::igptdecoderbatched::forward (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::igptdecoderbatched::forwardasync (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", false]], "tensorrt_llm::runtime::igptdecoderbatched::igptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched18IGptDecoderBatchedEv", false]], "tensorrt_llm::runtime::igptdecoderbatched::llmrequestptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13LlmRequestPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::requestvector (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13RequestVectorE", false]], "tensorrt_llm::runtime::igptdecoderbatched::setup (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::igptdecoderbatched::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched9TensorPtrE", false]], "tensorrt_llm::runtime::igptdecoderbatched::~igptdecoderbatched (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedD0Ev", false]], "tensorrt_llm::runtime::ipcmemory (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryE", false]], "tensorrt_llm::runtime::ipcmemory::allocateipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", false]], "tensorrt_llm::runtime::ipcmemory::bufferptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9BufferPtrE", false]], "tensorrt_llm::runtime::ipcmemory::destroyipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory16destroyIpcMemoryEv", false]], "tensorrt_llm::runtime::ipcmemory::flags_size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory10FLAGS_SIZEE", false]], "tensorrt_llm::runtime::ipcmemory::getcommptrs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9IpcMemory11getCommPtrsEv", false]], "tensorrt_llm::runtime::ipcmemory::ipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERK9IpcMemory", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERR9IpcMemory", false]], "tensorrt_llm::runtime::ipcmemory::mbuffer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mBufferE", false]], "tensorrt_llm::runtime::ipcmemory::mcommptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9mCommPtrsE", false]], "tensorrt_llm::runtime::ipcmemory::mopenipc (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory8mOpenIpcE", false]], "tensorrt_llm::runtime::ipcmemory::mtprank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mTpRankE", false]], "tensorrt_llm::runtime::ipcmemory::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERK9IpcMemory", false], [1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERR9IpcMemory", false]], "tensorrt_llm::runtime::ipcmemory::~ipcmemory (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryD0Ev", false]], "tensorrt_llm::runtime::ipcnvlsallocate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", false]], "tensorrt_llm::runtime::ipcnvlsfree (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", false]], "tensorrt_llm::runtime::ipcnvlshandle (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_handles (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle14ipc_uc_handlesE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_ptrs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle11ipc_uc_ptrsE", false]], "tensorrt_llm::runtime::ipcnvlshandle::ipc_uc_vas (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle10ipc_uc_vasE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9mc_handleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_ptr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6mc_ptrE", false]], "tensorrt_llm::runtime::ipcnvlshandle::mc_va (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5mc_vaE", false]], "tensorrt_llm::runtime::ipcnvlshandle::size (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle4sizeE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_handle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9uc_handleE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_ptr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6uc_ptrE", false]], "tensorrt_llm::runtime::ipcnvlshandle::uc_va (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5uc_vaE", false]], "tensorrt_llm::runtime::ipcnvlssupported (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime16ipcNvlsSupportedEv", false]], "tensorrt_llm::runtime::itensor (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensorE", false]], "tensorrt_llm::runtime::itensor::at (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", false]], "tensorrt_llm::runtime::itensor::castsize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", false]], "tensorrt_llm::runtime::itensor::dimtype64 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9DimType64E", false]], "tensorrt_llm::runtime::itensor::flattenn (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", false]], "tensorrt_llm::runtime::itensor::getdimension (c++ function)": [[1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", false]], "tensorrt_llm::runtime::itensor::getshape (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime7ITensor8getShapeEv", false]], "tensorrt_llm::runtime::itensor::itensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorERK7ITensor", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorEv", false]], "tensorrt_llm::runtime::itensor::makeshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", false]], "tensorrt_llm::runtime::itensor::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensoraSERK7ITensor", false]], "tensorrt_llm::runtime::itensor::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", false]], "tensorrt_llm::runtime::itensor::resize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", false]], "tensorrt_llm::runtime::itensor::shape (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor5ShapeE", false]], "tensorrt_llm::runtime::itensor::shapeequals (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", false], [1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", false], [1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", false], [1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", false]], "tensorrt_llm::runtime::itensor::sharedconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor14SharedConstPtrE", false]], "tensorrt_llm::runtime::itensor::sharedptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9SharedPtrE", false]], "tensorrt_llm::runtime::itensor::slice (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", false]], "tensorrt_llm::runtime::itensor::squeeze (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", false]], "tensorrt_llm::runtime::itensor::strides (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", false]], "tensorrt_llm::runtime::itensor::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9TensorMapE", false]], "tensorrt_llm::runtime::itensor::tostring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", false]], "tensorrt_llm::runtime::itensor::uniqueconstptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor14UniqueConstPtrE", false]], "tensorrt_llm::runtime::itensor::uniqueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9UniquePtrE", false]], "tensorrt_llm::runtime::itensor::unsqueeze (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", false]], "tensorrt_llm::runtime::itensor::view (c++ function)": [[1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", false]], "tensorrt_llm::runtime::itensor::volume (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", false]], "tensorrt_llm::runtime::itensor::volumenonnegative (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", false]], "tensorrt_llm::runtime::itensor::wrap (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", false], [1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", false], [1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", false]], "tensorrt_llm::runtime::itensor::~itensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime7ITensorD0Ev", false]], "tensorrt_llm::runtime::lamportinitializeall (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffersE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::generationlengths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers17generationLengthsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::lookaheaddecodingbuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::packedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11packedMasksE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::positionids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11positionIdsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::positionoffsets (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers15positionOffsetsE", false]], "tensorrt_llm::runtime::lookaheaddecodingbuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::lookaheadmodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModuleE", false]], "tensorrt_llm::runtime::lookaheadmodule::getexecutionconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime15LookaheadModule18getExecutionConfigEv", false]], "tensorrt_llm::runtime::lookaheadmodule::lookaheadmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleEv", false]], "tensorrt_llm::runtime::lookaheadmodule::mexecutionconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule16mExecutionConfigE", false]], "tensorrt_llm::runtime::lookaheadmodule::setexecutionconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffersE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::batchslotshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18batchSlotsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::cumsumlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers12cumSumLengthE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::disablelookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers24disableLookaheadDecodingEv", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::enablelookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23generationLengthsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21generationLengthsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::generationlengthshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers25generationLengthsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::insertinputtensors (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::lookaheadruntimebuffers (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmaskhost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers14packedMaskHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmaskhostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18packedMaskHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::packedmasksdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17packedMasksDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17positionIdsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15positionIdsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionidshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionIdsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetsdevice (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21positionOffsetsDeviceE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetshost (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionOffsetsHostE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::positionoffsetshostcopy (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23positionOffsetsHostCopyE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::reshape (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::setfrominputs (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::tensormap (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorMapE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorPtrE", false]], "tensorrt_llm::runtime::lookaheadruntimebuffers::usespecdecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15useSpecDecodingE", false]], "tensorrt_llm::runtime::loracache (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCacheE", false]], "tensorrt_llm::runtime::loracache::bump (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::bumptaskinprogress (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::claimpageswithevict (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", false]], "tensorrt_llm::runtime::loracache::copytask (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", false]], "tensorrt_llm::runtime::loracache::copytaskmappages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", false]], "tensorrt_llm::runtime::loracache::copytopages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", false]], "tensorrt_llm::runtime::loracache::determinenumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", false], [1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", false]], "tensorrt_llm::runtime::loracache::fits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", false]], "tensorrt_llm::runtime::loracache::get (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::getnumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache11getNumPagesEv", false]], "tensorrt_llm::runtime::loracache::getpageptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", false]], "tensorrt_llm::runtime::loracache::getstatus (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::has (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::isdone (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::isloaded (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::loadweights (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::loracache::loracache (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", false]], "tensorrt_llm::runtime::loracache::markalldone (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11markAllDoneEv", false]], "tensorrt_llm::runtime::loracache::marktaskdone (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", false]], "tensorrt_llm::runtime::loracache::mbuffermanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache14mBufferManagerE", false]], "tensorrt_llm::runtime::loracache::mcachemap (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9mCacheMapE", false]], "tensorrt_llm::runtime::loracache::mcachemutex (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mCacheMutexE", false]], "tensorrt_llm::runtime::loracache::mcachepagemanager (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mCachePageManagerE", false]], "tensorrt_llm::runtime::loracache::mdevicebuffermanagers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21mDeviceBufferManagersE", false]], "tensorrt_llm::runtime::loracache::mdonetasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10mDoneTasksE", false]], "tensorrt_llm::runtime::loracache::minprogresstasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16mInProgressTasksE", false]], "tensorrt_llm::runtime::loracache::mmodelconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mModelConfigE", false]], "tensorrt_llm::runtime::loracache::mmoduleidtomodule (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mModuleIdToModuleE", false]], "tensorrt_llm::runtime::loracache::mpagemanagerconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18mPageManagerConfigE", false]], "tensorrt_llm::runtime::loracache::mpagesmutex (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mPagesMutexE", false]], "tensorrt_llm::runtime::loracache::mworldconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mWorldConfigE", false]], "tensorrt_llm::runtime::loracache::put (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", false]], "tensorrt_llm::runtime::loracache::splittransposecpu (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracache::splittransposecpuinner (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracache::taskidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10TaskIdTypeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::adaptersize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig11adapterSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::insize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6inSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::layerid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7layerIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::moduleid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8moduleIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::numslots (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8numSlotsE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::outsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7outSizeE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::pageid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6pageIdE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::scalingvecpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17scalingVecPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::slotidx (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7slotIdxE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::tostring (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8toStringEv", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::weightsinpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig16weightsInPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfig::weightsoutpointer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17weightsOutPointerE", false]], "tensorrt_llm::runtime::loracache::tasklayermoduleconfiglistptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache28TaskLayerModuleConfigListPtrE", false]], "tensorrt_llm::runtime::loracache::taskvalue (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueE", false]], "tensorrt_llm::runtime::loracache::taskvalue::configs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7configsE", false]], "tensorrt_llm::runtime::loracache::taskvalue::done (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue4doneE", false]], "tensorrt_llm::runtime::loracache::taskvalue::inprogress (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue10inProgressE", false]], "tensorrt_llm::runtime::loracache::taskvalue::it (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue2itE", false]], "tensorrt_llm::runtime::loracache::taskvalue::loaded (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue6loadedE", false]], "tensorrt_llm::runtime::loracache::taskvalue::loadinprogress (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue14loadInProgressE", false]], "tensorrt_llm::runtime::loracache::taskvalue::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", false]], "tensorrt_llm::runtime::loracache::taskvalue::pageids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7pageIdsE", false]], "tensorrt_llm::runtime::loracache::taskvalue::taskvalue (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", false], [1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueEv", false]], "tensorrt_llm::runtime::loracache::taskvalue::~taskvalue (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueD0Ev", false]], "tensorrt_llm::runtime::loracache::taskvalueptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12TaskValuePtrE", false]], "tensorrt_llm::runtime::loracache::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TensorPtrE", false]], "tensorrt_llm::runtime::loracache::valuestatus (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatusE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_loaded (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_missing (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", false]], "tensorrt_llm::runtime::loracache::valuestatus::kvalue_status_processing (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", false]], "tensorrt_llm::runtime::loracachefullexception (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionE", false]], "tensorrt_llm::runtime::loracachefullexception::loracachefullexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", false]], "tensorrt_llm::runtime::loracachefullexception::~loracachefullexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionD0Ev", false]], "tensorrt_llm::runtime::loracachepagemanager (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManagerE", false]], "tensorrt_llm::runtime::loracachepagemanager::blockptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanager::claimpages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanager::initialize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", false]], "tensorrt_llm::runtime::loracachepagemanager::loracachepagemanager (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", false]], "tensorrt_llm::runtime::loracachepagemanager::mconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager7mConfigE", false]], "tensorrt_llm::runtime::loracachepagemanager::mfreepageids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12mFreePageIdsE", false]], "tensorrt_llm::runtime::loracachepagemanager::mispagefree (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mIsPageFreeE", false]], "tensorrt_llm::runtime::loracachepagemanager::mpageblocks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mPageBlocksE", false]], "tensorrt_llm::runtime::loracachepagemanager::mutablepageptr (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", false]], "tensorrt_llm::runtime::loracachepagemanager::numavailablepages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager17numAvailablePagesEv", false]], "tensorrt_llm::runtime::loracachepagemanager::pageptr (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", false]], "tensorrt_llm::runtime::loracachepagemanager::releasepages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", false]], "tensorrt_llm::runtime::loracachepagemanager::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager9TensorPtrE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfigE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig11getDataTypeEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getinittozero (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getInitToZeroEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getmaxpagesperblock (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig19getMaxPagesPerBlockEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getmemorytype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getMemoryTypeEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getnumcopystreams (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig17getNumCopyStreamsEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getpagewidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig12getPageWidthEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::getslotsperpage (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig15getSlotsPerPageEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::gettotalnumpages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig16getTotalNumPagesEv", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::loracachepagemanagerconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig9mDataTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::minittozero (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mInitToZeroE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mmaxpagesperblock (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17mMaxPagesPerBlockE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mmemorytype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mMemoryTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mnumcopystreams (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15mNumCopyStreamsE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mpagewidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig10mPageWidthE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mslotsperpage (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13mSlotsPerPageE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::mtotalnumpages (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig14mTotalNumPagesE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setdatatype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setinittozero (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setmaxpagesperblock (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setmemorytype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setnumcopystreams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setpagewidth (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::setslotsperpage (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", false]], "tensorrt_llm::runtime::loracachepagemanagerconfig::settotalnumpage (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", false]], "tensorrt_llm::runtime::loraexpectedexception (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionE", false]], "tensorrt_llm::runtime::loraexpectedexception::loraexpectedexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", false]], "tensorrt_llm::runtime::loraexpectedexception::~loraexpectedexception (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionD0Ev", false]], "tensorrt_llm::runtime::loramodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleE", false]], "tensorrt_llm::runtime::loramodule::createloramodules (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::flattenedinoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", false]], "tensorrt_llm::runtime::loramodule::indim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5inDimEv", false]], "tensorrt_llm::runtime::loramodule::indimfirst (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10inDimFirstEv", false]], "tensorrt_llm::runtime::loramodule::insize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::intpsplitdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12inTpSplitDimEv", false]], "tensorrt_llm::runtime::loramodule::localinadaptersize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localindim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::localinoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localinsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutadaptersize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::localoutsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::loramodule::localscalessize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", false]], "tensorrt_llm::runtime::loramodule::localtotalsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::loramodule::loramodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleEv", false]], "tensorrt_llm::runtime::loramodule::mindim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule6mInDimE", false]], "tensorrt_llm::runtime::loramodule::mindimfirst (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule11mInDimFirstE", false]], "tensorrt_llm::runtime::loramodule::mintpsplitdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule13mInTpSplitDimE", false]], "tensorrt_llm::runtime::loramodule::moduletype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleTypeE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_dense (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kATTN_DENSEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_k (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_KE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_q (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_QE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_qkv (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kATTN_QKVE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kattn_v (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_VE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_dense (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType17kCROSS_ATTN_DENSEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_k (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_KE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_q (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_QE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_qkv (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType15kCROSS_ATTN_QKVE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kcross_attn_v (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_VE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kinvalid (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType8kINVALIDE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_4h_to_h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_4H_TO_HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_gate (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMLP_GATEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_gate_up (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_GATE_UPE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_h_to_4h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_H_TO_4HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmlp_router (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMLP_ROUTERE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_4h_to_h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_4H_TO_HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_gate (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMOE_GATEE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_h_to_4h (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_H_TO_4HE", false]], "tensorrt_llm::runtime::loramodule::moduletype::kmoe_router (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMOE_ROUTERE", false]], "tensorrt_llm::runtime::loramodule::moutdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule7mOutDimE", false]], "tensorrt_llm::runtime::loramodule::moutdimfirst (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12mOutDimFirstE", false]], "tensorrt_llm::runtime::loramodule::mouttpsplitdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule14mOutTpSplitDimE", false]], "tensorrt_llm::runtime::loramodule::mtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule5mTypeE", false]], "tensorrt_llm::runtime::loramodule::name (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule4nameEv", false]], "tensorrt_llm::runtime::loramodule::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", false]], "tensorrt_llm::runtime::loramodule::outdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6outDimEv", false]], "tensorrt_llm::runtime::loramodule::outdimfirst (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11outDimFirstEv", false]], "tensorrt_llm::runtime::loramodule::outsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::outtpsplitdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule13outTpSplitDimEv", false]], "tensorrt_llm::runtime::loramodule::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule9TensorPtrE", false]], "tensorrt_llm::runtime::loramodule::tomodulename (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", false], [1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", false]], "tensorrt_llm::runtime::loramodule::tomoduletype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", false]], "tensorrt_llm::runtime::loramodule::value (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5valueEv", false]], "tensorrt_llm::runtime::lorataskidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14LoraTaskIdTypeE", false]], "tensorrt_llm::runtime::medusamodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModuleE", false]], "tensorrt_llm::runtime::medusamodule::getmedusachoices (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime12MedusaModule16getMedusaChoicesEv", false]], "tensorrt_llm::runtime::medusamodule::mdefaultmedusachoices (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule21mDefaultMedusaChoicesE", false]], "tensorrt_llm::runtime::medusamodule::medusachoices (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule13MedusaChoicesE", false]], "tensorrt_llm::runtime::medusamodule::medusamodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleEv", false]], "tensorrt_llm::runtime::medusamodule::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule9TensorPtrE", false]], "tensorrt_llm::runtime::memorycounters (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCountersE", false]], "tensorrt_llm::runtime::memorycounters::allocate (c++ function)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", false]], "tensorrt_llm::runtime::memorycounters::bytestostring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", false]], "tensorrt_llm::runtime::memorycounters::deallocate (c++ function)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", false]], "tensorrt_llm::runtime::memorycounters::difftype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8DiffTypeE", false]], "tensorrt_llm::runtime::memorycounters::getcpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getCpuEv", false]], "tensorrt_llm::runtime::memorycounters::getcpudiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getCpuDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getgpu (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getGpuEv", false]], "tensorrt_llm::runtime::memorycounters::getgpudiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getGpuDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getinstance (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11getInstanceEv", false]], "tensorrt_llm::runtime::memorycounters::getpinned (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters9getPinnedEv", false]], "tensorrt_llm::runtime::memorycounters::getpinneddiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getpinnedpool (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedPoolEv", false]], "tensorrt_llm::runtime::memorycounters::getpinnedpooldiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters17getPinnedPoolDiffEv", false]], "tensorrt_llm::runtime::memorycounters::getuvm (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getUVMEv", false]], "tensorrt_llm::runtime::memorycounters::getuvmdiff (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getUVMDiffEv", false]], "tensorrt_llm::runtime::memorycounters::mcpu (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mCpuE", false]], "tensorrt_llm::runtime::memorycounters::mcpudiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mCpuDiffE", false]], "tensorrt_llm::runtime::memorycounters::memorycounters (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters14MemoryCountersEv", false]], "tensorrt_llm::runtime::memorycounters::mgpu (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mGpuE", false]], "tensorrt_llm::runtime::memorycounters::mgpudiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mGpuDiffE", false]], "tensorrt_llm::runtime::memorycounters::mpinned (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters7mPinnedE", false]], "tensorrt_llm::runtime::memorycounters::mpinneddiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedDiffE", false]], "tensorrt_llm::runtime::memorycounters::mpinnedpool (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedPoolE", false]], "tensorrt_llm::runtime::memorycounters::mpinnedpooldiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters15mPinnedPoolDiffE", false]], "tensorrt_llm::runtime::memorycounters::muvm (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mUVME", false]], "tensorrt_llm::runtime::memorycounters::muvmdiff (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mUVMDiffE", false]], "tensorrt_llm::runtime::memorycounters::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10SizeType32E", false]], "tensorrt_llm::runtime::memorycounters::tostring (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters8toStringEv", false]], "tensorrt_llm::runtime::memorytype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryTypeE", false]], "tensorrt_llm::runtime::memorytype::kcpu (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kCPUE", false]], "tensorrt_llm::runtime::memorytype::kgpu (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kGPUE", false]], "tensorrt_llm::runtime::memorytype::kpinned (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType7kPINNEDE", false]], "tensorrt_llm::runtime::memorytype::kpinnedpool (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType11kPINNEDPOOLE", false]], "tensorrt_llm::runtime::memorytype::kuvm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kUVME", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEE5valueE", false]], "tensorrt_llm::runtime::memorytypestring (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEEE", false]], "tensorrt_llm::runtime::memorytypestring::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEE5valueE", false]], "tensorrt_llm::runtime::modelconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfigE", false]], "tensorrt_llm::runtime::modelconfig::computecontextlogits (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEv", false]], "tensorrt_llm::runtime::modelconfig::computegenerationlogits (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEv", false]], "tensorrt_llm::runtime::modelconfig::countlocallayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::countlowerranklayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::disableseamlesslookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig32disableSeamlessLookaheadDecodingEv", false]], "tensorrt_llm::runtime::modelconfig::enableseamlesslookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getcontextfmha (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getContextFMHAEv", false]], "tensorrt_llm::runtime::modelconfig::getdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getDataTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getencoderhiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getEncoderHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getfirstlocallayer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getFirstLocalLayerE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getgemmallreducedtype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getGemmAllReduceDtypeEv", false]], "tensorrt_llm::runtime::modelconfig::gethiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getkvcachetype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getKVCacheTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getkvdatatype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getKvDataTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getlayertypes (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getLayerTypesEv", false]], "tensorrt_llm::runtime::modelconfig::getlogitsdtype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLogitsDtypeEv", false]], "tensorrt_llm::runtime::modelconfig::getloramodules (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLoraModulesEv", false]], "tensorrt_llm::runtime::modelconfig::getmanageweightstype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getManageWeightsTypeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxbatchsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBatchSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxdecodingdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig25getMaxDecodingDraftTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxdecodingtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getMaxDecodingTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxencoderlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMaxEncoderLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxinputlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxInputLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxlorarank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxLoraRankEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxnumtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxNumTokensEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxpositionembeddings (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24getMaxPositionEmbeddingsEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxpromptembeddingtablesize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig30getMaxPromptEmbeddingTableSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmaxsequencelen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getMaxSequenceLenEv", false]], "tensorrt_llm::runtime::modelconfig::getmlphiddensize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMlpHiddenSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getmodelname (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getModelNameEv", false]], "tensorrt_llm::runtime::modelconfig::getmodelvariant (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getModelVariantEv", false]], "tensorrt_llm::runtime::modelconfig::getnbattentionlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnbheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10getNbHeadsEv", false]], "tensorrt_llm::runtime::modelconfig::getnbkvheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnblayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnbrnnlayers (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", false]], "tensorrt_llm::runtime::modelconfig::getnumkvheadsperlayer (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getNumKvHeadsPerLayerEv", false]], "tensorrt_llm::runtime::modelconfig::getnumkvheadsperlayerlocalrange (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::modelconfig::getnumlanguages (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getNumLanguagesEv", false]], "tensorrt_llm::runtime::modelconfig::getoptprofilessplitpoints (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig25getOptProfilesSplitPointsEv", false]], "tensorrt_llm::runtime::modelconfig::getpagedcontextfmha (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19getPagedContextFMHAEv", false]], "tensorrt_llm::runtime::modelconfig::getppreducescatter (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getPpReduceScatterEv", false]], "tensorrt_llm::runtime::modelconfig::getquantmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getQuantModeEv", false]], "tensorrt_llm::runtime::modelconfig::getrnnconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getRnnConfigEv", false]], "tensorrt_llm::runtime::modelconfig::getrotaryembeddingdim (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getRotaryEmbeddingDimEv", false]], "tensorrt_llm::runtime::modelconfig::getsizeperhead (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getSizePerHeadEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig26getSpeculativeDecodingModeEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmodule (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28getSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::getspeculativedecodingmoduleptr (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", false]], "tensorrt_llm::runtime::modelconfig::getsumlocalkvheads (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", false]], "tensorrt_llm::runtime::modelconfig::gettokensperblock (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getTokensPerBlockEv", false]], "tensorrt_llm::runtime::modelconfig::getvocabsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getVocabSizeEv", false]], "tensorrt_llm::runtime::modelconfig::getvocabsizepadded (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::hasrnnconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12hasRnnConfigEv", false]], "tensorrt_llm::runtime::modelconfig::hasspeculativedecodingmodule (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28hasSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::iscontinuouskvcache (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19isContinuousKVCacheEv", false]], "tensorrt_llm::runtime::modelconfig::iskvcacheenabled (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16isKVCacheEnabledEv", false]], "tensorrt_llm::runtime::modelconfig::ismultimodal (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12isMultiModalEv", false]], "tensorrt_llm::runtime::modelconfig::ispagedkvcache (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14isPagedKVCacheEv", false]], "tensorrt_llm::runtime::modelconfig::isrnnbased (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10isRnnBasedEv", false]], "tensorrt_llm::runtime::modelconfig::istransformerbased (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18isTransformerBasedEv", false]], "tensorrt_llm::runtime::modelconfig::iswhisper (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig9isWhisperEv", false]], "tensorrt_llm::runtime::modelconfig::kdefault_num_tokens_per_block (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig29kDEFAULT_NUM_TOKENS_PER_BLOCKE", false]], "tensorrt_llm::runtime::modelconfig::kopt_profiles_split_points (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26kOPT_PROFILES_SPLIT_POINTSE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheTypeE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kcontinuous (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType11kCONTINUOUSE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kdisabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType9kDISABLEDE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetype::kpaged (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType6kPAGEDE", false]], "tensorrt_llm::runtime::modelconfig::kvcachetypefromstring (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", false]], "tensorrt_llm::runtime::modelconfig::layertype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerTypeE", false]], "tensorrt_llm::runtime::modelconfig::layertype::kattention (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kATTENTIONE", false]], "tensorrt_llm::runtime::modelconfig::layertype::klinear (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType7kLINEARE", false]], "tensorrt_llm::runtime::modelconfig::layertype::knoop (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType5kNOOPE", false]], "tensorrt_llm::runtime::modelconfig::layertype::krecurrent (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kRECURRENTE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsTypeE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype::kdisabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType9kDisabledE", false]], "tensorrt_llm::runtime::modelconfig::manageweightstype::kenabled (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType8kEnabledE", false]], "tensorrt_llm::runtime::modelconfig::mcomputecontextlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mComputeContextLogitsE", false]], "tensorrt_llm::runtime::modelconfig::mcomputegenerationlogits (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mComputeGenerationLogitsE", false]], "tensorrt_llm::runtime::modelconfig::mcontextfmha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mContextFMHAE", false]], "tensorrt_llm::runtime::modelconfig::mdatatype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mDataTypeE", false]], "tensorrt_llm::runtime::modelconfig::mencoderhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mEncoderHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::mgemmallreducedtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mGemmAllReduceDtypeE", false]], "tensorrt_llm::runtime::modelconfig::mhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::minputpacked (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mInputPackedE", false]], "tensorrt_llm::runtime::modelconfig::mkvcachetype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mKVCacheTypeE", false]], "tensorrt_llm::runtime::modelconfig::mlayertypes (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mLayerTypesE", false]], "tensorrt_llm::runtime::modelconfig::mlogitsdtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLogitsDtypeE", false]], "tensorrt_llm::runtime::modelconfig::mloramodules (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLoraModulesE", false]], "tensorrt_llm::runtime::modelconfig::mmanageweightstype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mManageWeightsTypeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxbatchsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBatchSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxbeamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBeamWidthE", false]], "tensorrt_llm::runtime::modelconfig::mmaxencoderlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMaxEncoderLenE", false]], "tensorrt_llm::runtime::modelconfig::mmaxinputlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxInputLenE", false]], "tensorrt_llm::runtime::modelconfig::mmaxlorarank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxLoraRankE", false]], "tensorrt_llm::runtime::modelconfig::mmaxnumtokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxNumTokensE", false]], "tensorrt_llm::runtime::modelconfig::mmaxpositionembeddings (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mMaxPositionEmbeddingsE", false]], "tensorrt_llm::runtime::modelconfig::mmaxpromptembeddingtablesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mMaxPromptEmbeddingTableSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmaxsequencelen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mMaxSequenceLenE", false]], "tensorrt_llm::runtime::modelconfig::mmlphiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMlpHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::mmodelname (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mModelNameE", false]], "tensorrt_llm::runtime::modelconfig::mmodelvariant (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mModelVariantE", false]], "tensorrt_llm::runtime::modelconfig::mnbattentionlayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mNbAttentionLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnbheads (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig8mNbHeadsE", false]], "tensorrt_llm::runtime::modelconfig::mnblayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mNbLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnbrnnlayers (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mNbRnnLayersE", false]], "tensorrt_llm::runtime::modelconfig::mnumkvheadsperattentionlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mNumKvHeadsPerAttentionLayerE", false]], "tensorrt_llm::runtime::modelconfig::mnumkvheadspercrossattentionlayer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig33mNumKvHeadsPerCrossAttentionLayerE", false]], "tensorrt_llm::runtime::modelconfig::mnumlanguages (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mNumLanguagesE", false]], "tensorrt_llm::runtime::modelconfig::modelconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariantE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kchatglm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant8kChatGlmE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kencdec (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant7kEncDecE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kglm (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGlmE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kgpt (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGptE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::kmamba (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant6kMambaE", false]], "tensorrt_llm::runtime::modelconfig::modelvariant::krecurrentgemma (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant15kRecurrentGemmaE", false]], "tensorrt_llm::runtime::modelconfig::mpagedcontextfmha (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17mPagedContextFMHAE", false]], "tensorrt_llm::runtime::modelconfig::mpagedstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mPagedStateE", false]], "tensorrt_llm::runtime::modelconfig::mppreducescatter (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16mPpReduceScatterE", false]], "tensorrt_llm::runtime::modelconfig::mquantmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mQuantModeE", false]], "tensorrt_llm::runtime::modelconfig::mrnnconfig (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mRnnConfigE", false]], "tensorrt_llm::runtime::modelconfig::mrotaryembeddingdim (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mRotaryEmbeddingDimE", false]], "tensorrt_llm::runtime::modelconfig::msizeperhead (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mSizePerHeadE", false]], "tensorrt_llm::runtime::modelconfig::mskipcrossattnblocks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20mSkipCrossAttnBlocksE", false]], "tensorrt_llm::runtime::modelconfig::mspeculativedecodingmode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mSpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::modelconfig::mspeculativedecodingmodule (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26mSpeculativeDecodingModuleE", false]], "tensorrt_llm::runtime::modelconfig::mtokensperblock (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mTokensPerBlockE", false]], "tensorrt_llm::runtime::modelconfig::musecrossattention (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseCrossAttentionE", false]], "tensorrt_llm::runtime::modelconfig::musegemmallreduceplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23mUseGemmAllReducePluginE", false]], "tensorrt_llm::runtime::modelconfig::musegptattentionplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseGptAttentionPluginE", false]], "tensorrt_llm::runtime::modelconfig::museloraplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mUseLoraPluginE", false]], "tensorrt_llm::runtime::modelconfig::musemambaconv1dplugin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUseMambaConv1dPluginE", false]], "tensorrt_llm::runtime::modelconfig::musemrope (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mUseMropeE", false]], "tensorrt_llm::runtime::modelconfig::musepositionembedding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUsePositionEmbeddingE", false]], "tensorrt_llm::runtime::modelconfig::museshapeinference (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseShapeInferenceE", false]], "tensorrt_llm::runtime::modelconfig::musetokentypeembedding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseTokenTypeEmbeddingE", false]], "tensorrt_llm::runtime::modelconfig::mvocabsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mVocabSizeE", false]], "tensorrt_llm::runtime::modelconfig::resetspeculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30resetSpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfigE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::convkernel (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig10convKernelE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnconvdimsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig14rnnConvDimSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnheadsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig11rnnHeadSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::rnnhiddensize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig13rnnHiddenSizeE", false]], "tensorrt_llm::runtime::modelconfig::rnnconfig::statesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig9stateSizeE", false]], "tensorrt_llm::runtime::modelconfig::setcontextfmha (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", false]], "tensorrt_llm::runtime::modelconfig::setencoderhiddensize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setgemmallreducedtype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::setkvcachetype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", false]], "tensorrt_llm::runtime::modelconfig::setlayertypes (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", false]], "tensorrt_llm::runtime::modelconfig::setlogitsdtype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", false]], "tensorrt_llm::runtime::modelconfig::setloramodules (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", false]], "tensorrt_llm::runtime::modelconfig::setmanageweightstype (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", false]], "tensorrt_llm::runtime::modelconfig::setmaxbatchsize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxbeamwidth (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxencoderlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxinputlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxlorarank (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxnumtokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setmaxpositionembeddings (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxpromptembeddingtablesize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmaxsequencelen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmlphiddensize (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setmodelname (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", false]], "tensorrt_llm::runtime::modelconfig::setmodelvariant (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", false]], "tensorrt_llm::runtime::modelconfig::setnbcrosskvheads (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setnbkvheads (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setnumkvheadspercrosslayer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setnumkvheadsperlayer (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setnumlanguages (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", false]], "tensorrt_llm::runtime::modelconfig::setpagedcontextfmha (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", false]], "tensorrt_llm::runtime::modelconfig::setppreducescatter (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", false]], "tensorrt_llm::runtime::modelconfig::setquantmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", false]], "tensorrt_llm::runtime::modelconfig::setrnnconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", false]], "tensorrt_llm::runtime::modelconfig::setrotaryembeddingdim (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setsizeperhead (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setskipcrossattnblocks (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", false]], "tensorrt_llm::runtime::modelconfig::setspeculativedecodingmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", false]], "tensorrt_llm::runtime::modelconfig::setspeculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", false]], "tensorrt_llm::runtime::modelconfig::settokensperblock (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", false]], "tensorrt_llm::runtime::modelconfig::setusecrossattention (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", false]], "tensorrt_llm::runtime::modelconfig::setusemrope (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", false]], "tensorrt_llm::runtime::modelconfig::setusepositionembedding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", false]], "tensorrt_llm::runtime::modelconfig::setuseshapeinference (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", false]], "tensorrt_llm::runtime::modelconfig::setusetokentypeembedding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", false]], "tensorrt_llm::runtime::modelconfig::skipcrossattnblocks (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19skipCrossAttnBlocksEv", false]], "tensorrt_llm::runtime::modelconfig::supportsinflightbatching (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24supportsInflightBatchingEv", false]], "tensorrt_llm::runtime::modelconfig::usecrossattention (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useCrossAttentionEv", false]], "tensorrt_llm::runtime::modelconfig::usegemmallreduceplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEv", false]], "tensorrt_llm::runtime::modelconfig::usegptattentionplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEv", false]], "tensorrt_llm::runtime::modelconfig::uselanguageadapter (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18useLanguageAdapterEv", false]], "tensorrt_llm::runtime::modelconfig::useloraplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13useLoraPluginEv", false]], "tensorrt_llm::runtime::modelconfig::usemambaconv1dplugin (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEv", false]], "tensorrt_llm::runtime::modelconfig::usemrope (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig8useMropeEv", false]], "tensorrt_llm::runtime::modelconfig::usepackedinput (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14usePackedInputEv", false]], "tensorrt_llm::runtime::modelconfig::usepagedstate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", false], [1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13usePagedStateEv", false]], "tensorrt_llm::runtime::modelconfig::usepositionembedding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20usePositionEmbeddingEv", false]], "tensorrt_llm::runtime::modelconfig::useprompttuning (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15usePromptTuningEv", false]], "tensorrt_llm::runtime::modelconfig::useshapeinference (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useShapeInferenceEv", false]], "tensorrt_llm::runtime::modelconfig::usetokentypeembedding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useTokenTypeEmbeddingEv", false]], "tensorrt_llm::runtime::mpi_group_barrier (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", false]], "tensorrt_llm::runtime::operator<< (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", false], [1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::pointerelementtype (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", false]], "tensorrt_llm::runtime::prompttuningparams (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParamsE", false]], "tensorrt_llm::runtime::prompttuningparams::filltaskstensor (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", false]], "tensorrt_llm::runtime::prompttuningparams::prompttuningparams (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", false]], "tensorrt_llm::runtime::prompttuningparams::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams10SizeType32E", false]], "tensorrt_llm::runtime::prompttuningparams::tensorptr (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams9TensorPtrE", false]], "tensorrt_llm::runtime::rawengine (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngineE", false]], "tensorrt_llm::runtime::rawengine::getaddress (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getAddressEv", false]], "tensorrt_llm::runtime::rawengine::gethostmemory (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine13getHostMemoryEv", false]], "tensorrt_llm::runtime::rawengine::getmanagedweightsmapopt (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine23getManagedWeightsMapOptEv", false]], "tensorrt_llm::runtime::rawengine::getpath (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getPathEv", false]], "tensorrt_llm::runtime::rawengine::getpathopt (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getPathOptEv", false]], "tensorrt_llm::runtime::rawengine::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getSizeEv", false]], "tensorrt_llm::runtime::rawengine::gettype (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getTypeEv", false]], "tensorrt_llm::runtime::rawengine::mengineaddr (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineAddrE", false]], "tensorrt_llm::runtime::rawengine::menginebuffer (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine13mEngineBufferE", false]], "tensorrt_llm::runtime::rawengine::menginepath (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEnginePathE", false]], "tensorrt_llm::runtime::rawengine::menginesize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineSizeE", false]], "tensorrt_llm::runtime::rawengine::mmanagedweightsmap (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine18mManagedWeightsMapE", false]], "tensorrt_llm::runtime::rawengine::mtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine5mTypeE", false]], "tensorrt_llm::runtime::rawengine::rawengine (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", false], [1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", false], [1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", false]], "tensorrt_llm::runtime::rawengine::setmanagedweightsmap (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", false]], "tensorrt_llm::runtime::rawengine::setpath (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", false]], "tensorrt_llm::runtime::rawengine::type (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4TypeE", false]], "tensorrt_llm::runtime::rawengine::type::addresswithsize (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", false]], "tensorrt_llm::runtime::rawengine::type::filepath (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", false]], "tensorrt_llm::runtime::rawengine::type::hostmemory (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", false]], "tensorrt_llm::runtime::requesttype (c++ enum)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestTypeE", false]], "tensorrt_llm::runtime::requesttype::kcontext (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestType8kCONTEXTE", false]], "tensorrt_llm::runtime::requesttype::kgeneration (c++ enumerator)": [[1, "_CPPv4N12tensorrt_llm7runtime11RequestType11kGENERATIONE", false]], "tensorrt_llm::runtime::runtimedefaults (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaultsE", false]], "tensorrt_llm::runtime::runtimedefaults::maxattentionwindowvec (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults21maxAttentionWindowVecE", false]], "tensorrt_llm::runtime::runtimedefaults::runtimedefaults (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", false], [1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsEv", false]], "tensorrt_llm::runtime::runtimedefaults::sinktokenlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15sinkTokenLengthE", false]], "tensorrt_llm::runtime::samplingconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfigE", false]], "tensorrt_llm::runtime::samplingconfig::beamsearchdiversityrate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig23beamSearchDiversityRateE", false]], "tensorrt_llm::runtime::samplingconfig::beamwidth (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9beamWidthE", false]], "tensorrt_llm::runtime::samplingconfig::beamwidtharray (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14beamWidthArrayE", false]], "tensorrt_llm::runtime::samplingconfig::cumlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11cumLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::draftacceptancethreshold (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig24draftAcceptanceThresholdE", false]], "tensorrt_llm::runtime::samplingconfig::earlystopping (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13earlyStoppingE", false]], "tensorrt_llm::runtime::samplingconfig::floattype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9FloatTypeE", false]], "tensorrt_llm::runtime::samplingconfig::frequencypenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig16frequencyPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::fusevalues (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", false]], "tensorrt_llm::runtime::samplingconfig::getmaxbeamwidth (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig15getMaxBeamWidthEv", false]], "tensorrt_llm::runtime::samplingconfig::getnumreturnbeams (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig17getNumReturnBeamsEv", false]], "tensorrt_llm::runtime::samplingconfig::lengthpenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13lengthPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::minlength (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9minLengthE", false]], "tensorrt_llm::runtime::samplingconfig::minp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4minPE", false]], "tensorrt_llm::runtime::samplingconfig::norepeatngramsize (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17noRepeatNgramSizeE", false]], "tensorrt_llm::runtime::samplingconfig::normalizelogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17normalizeLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::numreturnsequences (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig18numReturnSequencesE", false]], "tensorrt_llm::runtime::samplingconfig::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", false]], "tensorrt_llm::runtime::samplingconfig::optvec (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", false]], "tensorrt_llm::runtime::samplingconfig::originaltemperature (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig19originalTemperatureE", false]], "tensorrt_llm::runtime::samplingconfig::outputlogprobs (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14outputLogProbsE", false]], "tensorrt_llm::runtime::samplingconfig::presencepenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15presencePenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::randomseed (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig10randomSeedE", false]], "tensorrt_llm::runtime::samplingconfig::repetitionpenalty (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17repetitionPenaltyE", false]], "tensorrt_llm::runtime::samplingconfig::samplingconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", false], [1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", false]], "tensorrt_llm::runtime::samplingconfig::temperature (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11temperatureE", false]], "tensorrt_llm::runtime::samplingconfig::topk (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topKE", false]], "tensorrt_llm::runtime::samplingconfig::topkmedusaheads (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15topKMedusaHeadsE", false]], "tensorrt_llm::runtime::samplingconfig::topp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topPE", false]], "tensorrt_llm::runtime::samplingconfig::toppdecay (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9topPDecayE", false]], "tensorrt_llm::runtime::samplingconfig::toppmin (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig7topPMinE", false]], "tensorrt_llm::runtime::samplingconfig::toppresetids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig12topPResetIdsE", false]], "tensorrt_llm::runtime::samplingconfig::usedefaultvalues (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", false]], "tensorrt_llm::runtime::samplingconfig::validate (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig8validateEv", false]], "tensorrt_llm::runtime::samplingconfig::validatevec (c++ function)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", false]], "tensorrt_llm::runtime::sizetype32 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10SizeType32E", false]], "tensorrt_llm::runtime::sizetype64 (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime10SizeType64E", false]], "tensorrt_llm::runtime::speculativedecodingmode (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingModeE", false]], "tensorrt_llm::runtime::speculativedecodingmode::allbitset (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::anybitset (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::drafttokensexternal (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19DraftTokensExternalEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::eagle (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5EagleEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::explicitdrafttokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19ExplicitDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::hasdraftlogits (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode14hasDraftLogitsEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isdrafttokensexternal (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isDraftTokensExternalEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::iseagle (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode7isEagleEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isexplicitdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isExplicitDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::islookaheaddecoding (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19isLookaheadDecodingEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::ismedusa (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode8isMedusaEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::isnone (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode6isNoneEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::kdrafttokensexternal (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kDraftTokensExternalE", false]], "tensorrt_llm::runtime::speculativedecodingmode::keagle (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6kEagleE", false]], "tensorrt_llm::runtime::speculativedecodingmode::kexplicitdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kExplicitDraftTokensE", false]], "tensorrt_llm::runtime::speculativedecodingmode::klookaheaddecoding (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode18kLookaheadDecodingE", false]], "tensorrt_llm::runtime::speculativedecodingmode::kmedusa (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode7kMedusaE", false]], "tensorrt_llm::runtime::speculativedecodingmode::knone (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5kNoneE", false]], "tensorrt_llm::runtime::speculativedecodingmode::lookaheaddecoding (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode17LookaheadDecodingEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::medusa (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6MedusaEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::mstate (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6mStateE", false]], "tensorrt_llm::runtime::speculativedecodingmode::needsdecoderprologue (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode20needsDecoderPrologueEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::needskvcacherewind (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18needsKVCacheRewindEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::none (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode4NoneEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", false]], "tensorrt_llm::runtime::speculativedecodingmode::predictsdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19predictsDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::requiresattentionmask (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21requiresAttentionMaskEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::speculativedecodingmode (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", false]], "tensorrt_llm::runtime::speculativedecodingmode::underlyingtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode14UnderlyingTypeE", false]], "tensorrt_llm::runtime::speculativedecodingmode::updatespositionids (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18updatesPositionIdsEv", false]], "tensorrt_llm::runtime::speculativedecodingmode::variabledraftlength (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19variableDraftLengthEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::computenumpackedmasks (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule21computeNumPackedMasksEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdecodingdrafttokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule25getMaxDecodingDraftTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdecodingtokens (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule20getMaxDecodingTokensEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxdraftpathlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule18getMaxDraftPathLenEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxnumpaths (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule14getMaxNumPathsEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getmaxpathlen (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule13getMaxPathLenEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::getnumpackedmasks (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule17getNumPackedMasksEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxdecodingdrafttokens (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule23mMaxDecodingDraftTokensE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxdraftpathlen (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule16mMaxDraftPathLenE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxnumpackedmasks (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18mMaxNumPackedMasksE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::mmaxnumpaths (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule12mMaxNumPathsE", false]], "tensorrt_llm::runtime::speculativedecodingmodule::operator= (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxdraftpathlen (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxdrafttokens (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::setmaxnumpaths (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", false]], "tensorrt_llm::runtime::speculativedecodingmodule::speculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", false], [1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", false], [1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleEv", false]], "tensorrt_llm::runtime::speculativedecodingmodule::~speculativedecodingmodule (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleD0Ev", false]], "tensorrt_llm::runtime::stringptrmap (c++ type)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", false]], "tensorrt_llm::runtime::tllmlogger (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLoggerE", false]], "tensorrt_llm::runtime::tllmlogger::getlevel (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8getLevelEv", false]], "tensorrt_llm::runtime::tllmlogger::log (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", false]], "tensorrt_llm::runtime::tllmlogger::setlevel (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", false]], "tensorrt_llm::runtime::to_string (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", false], [1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", false]], "tensorrt_llm::runtime::tokenextraidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16TokenExtraIdTypeE", false]], "tensorrt_llm::runtime::tokenidtype (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime11TokenIdTypeE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIbEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIbE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIfEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIfE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeI4halfEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeI4halfE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", false]], "tensorrt_llm::runtime::trtdatatype::kunderlyingtype (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE15kUnderlyingTypeE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE5valueE", false]], "tensorrt_llm::runtime::trtdatatype (c++ struct)": [[1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIPvEE", false]], "tensorrt_llm::runtime::trtdatatype::value (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIPvE5valueE", false]], "tensorrt_llm::runtime::uniquetoken (c++ struct)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueTokenE", false]], "tensorrt_llm::runtime::uniquetoken::operator== (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", false]], "tensorrt_llm::runtime::uniquetoken::tokenextraid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken12tokenExtraIdE", false]], "tensorrt_llm::runtime::uniquetoken::tokenid (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken7tokenIdE", false]], "tensorrt_llm::runtime::vectokenextraids (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime16VecTokenExtraIdsE", false]], "tensorrt_llm::runtime::vecuniquetokens (c++ type)": [[1, "_CPPv4N12tensorrt_llm7runtime15VecUniqueTokensE", false]], "tensorrt_llm::runtime::worldconfig (c++ class)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfigE", false]], "tensorrt_llm::runtime::worldconfig::enableattentiondp (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17enableAttentionDPEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getContextParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getContextParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::getcontextparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getContextParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::getdevice (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig9getDeviceEv", false]], "tensorrt_llm::runtime::worldconfig::getdeviceof (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", false]], "tensorrt_llm::runtime::worldconfig::getgpuspergroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig15getGpusPerGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getgpuspernode (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14getGpusPerNodeEv", false]], "tensorrt_llm::runtime::worldconfig::getlastrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getLastRankEv", false]], "tensorrt_llm::runtime::worldconfig::getlocalrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig12getLocalRankEv", false]], "tensorrt_llm::runtime::worldconfig::getnoderank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getNodeRankEv", false]], "tensorrt_llm::runtime::worldconfig::getnoderankof (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig24getPipelineParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getPipelineParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::getpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::getrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getRankEv", false]], "tensorrt_llm::runtime::worldconfig::getsize (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getSizeEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelgroup (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getTensorParallelGroupEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelism (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig20getTensorParallelismEv", false]], "tensorrt_llm::runtime::worldconfig::gettensorparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getTensorParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::iscontextparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17isContextParallelEv", false]], "tensorrt_llm::runtime::worldconfig::isfirstcontextparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isFirstContextParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::isfirstpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig27isFirstPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::isfirsttensorparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig25isFirstTensorParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::islastpipelineparallelrank (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isLastPipelineParallelRankEv", false]], "tensorrt_llm::runtime::worldconfig::ispipelineparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig18isPipelineParallelEv", false]], "tensorrt_llm::runtime::worldconfig::istensorparallel (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig16isTensorParallelEv", false]], "tensorrt_llm::runtime::worldconfig::kdefaultgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19kDefaultGpusPerNodeE", false]], "tensorrt_llm::runtime::worldconfig::mcontextparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19mContextParallelismE", false]], "tensorrt_llm::runtime::worldconfig::mdeviceids (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig10mDeviceIdsE", false]], "tensorrt_llm::runtime::worldconfig::menableattentiondp (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mEnableAttentionDPE", false]], "tensorrt_llm::runtime::worldconfig::mgpuspernode (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig12mGpusPerNodeE", false]], "tensorrt_llm::runtime::worldconfig::mpi (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", false]], "tensorrt_llm::runtime::worldconfig::mpipelineparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig20mPipelineParallelismE", false]], "tensorrt_llm::runtime::worldconfig::mrank (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig5mRankE", false]], "tensorrt_llm::runtime::worldconfig::mtensorparallelism (c++ member)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mTensorParallelismE", false]], "tensorrt_llm::runtime::worldconfig::validmpiconfig (c++ function)": [[1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14validMpiConfigEv", false]], "tensorrt_llm::runtime::worldconfig::worldconfig (c++ function)": [[1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", false]], "text (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.text", false]], "text_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.text_diff", false]], "text_diff (tensorrt_llm.llmapi.completionoutput property)": [[70, "id4", false]], "timestepembedding (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.TimestepEmbedding", false]], "timesteps (class in tensorrt_llm.layers.embedding)": [[83, "tensorrt_llm.layers.embedding.Timesteps", false]], "to_dict() (tensorrt_llm.llmapi.buildconfig method)": [[70, "tensorrt_llm.llmapi.BuildConfig.to_dict", false]], "to_dict() (tensorrt_llm.llmapi.calibconfig method)": [[70, "tensorrt_llm.llmapi.CalibConfig.to_dict", false]], "to_dict() (tensorrt_llm.llmapi.quantconfig method)": [[70, "tensorrt_llm.llmapi.QuantConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.chatglmconfig method)": [[84, "tensorrt_llm.models.ChatGLMConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.cogvlmconfig method)": [[84, "tensorrt_llm.models.CogVLMConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.dbrxconfig method)": [[84, "tensorrt_llm.models.DbrxConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.falconconfig method)": [[84, "tensorrt_llm.models.FalconConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gemmaconfig method)": [[84, "tensorrt_llm.models.GemmaConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gptconfig method)": [[84, "tensorrt_llm.models.GPTConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.gptjconfig method)": [[84, "tensorrt_llm.models.GPTJConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.llamaconfig method)": [[84, "tensorrt_llm.models.LLaMAConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.medusaconfig method)": [[84, "tensorrt_llm.models.MedusaConfig.to_dict", false]], "to_dict() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.to_dict", false]], "to_json_file() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.to_json_file", false]], "to_layer_quant_config() (tensorrt_llm.models.pretrainedconfig method)": [[84, "tensorrt_llm.models.PretrainedConfig.to_layer_quant_config", false]], "to_legacy_setting() (tensorrt_llm.plugin.pluginconfig method)": [[85, "tensorrt_llm.plugin.PluginConfig.to_legacy_setting", false]], "token_drop() (tensorrt_llm.layers.embedding.labelembedding method)": [[83, "tensorrt_llm.layers.embedding.LabelEmbedding.token_drop", false]], "token_end (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.token_end", false]], "token_ids (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.token_ids", false]], "token_ids_diff (tensorrt_llm.llmapi.completionoutput attribute)": [[70, "tensorrt_llm.llmapi.CompletionOutput.token_ids_diff", false]], "token_ids_diff (tensorrt_llm.llmapi.completionoutput property)": [[70, "id5", false]], "token_range_retention_configs (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.token_range_retention_configs", false]], "token_start (tensorrt_llm.llmapi.kvcacheretentionconfig.tokenrangeretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig.token_start", false]], "tokenizer (tensorrt_llm.llmapi.llm attribute)": [[70, "tensorrt_llm.llmapi.LLM.tokenizer", false]], "tokenizer (tensorrt_llm.llmapi.llm property)": [[70, "id0", false]], "tokenizer_image_token() (tensorrt_llm.runtime.multimodalmodelrunner static method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.tokenizer_image_token", false]], "tokenizer_max_seq_length (tensorrt_llm.llmapi.calibconfig attribute)": [[70, "tensorrt_llm.llmapi.CalibConfig.tokenizer_max_seq_length", false]], "tokens_per_block (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.tokens_per_block", false]], "tokens_per_block (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.tokens_per_block", false]], "top_k (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.top_k", false]], "top_k (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.top_k", false]], "top_p (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.top_p", false]], "top_p (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.top_p", false]], "top_p_decay (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.top_p_decay", false]], "top_p_decay (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.top_p_decay", false]], "top_p_min (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.top_p_min", false]], "top_p_min (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.top_p_min", false]], "top_p_reset_ids (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.top_p_reset_ids", false]], "top_p_reset_ids (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.top_p_reset_ids", false]], "topk() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.topk", false]], "torch_compile_enable_userbuffers (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.torch_compile_enable_userbuffers", false]], "torch_compile_enabled (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.torch_compile_enabled", false]], "torch_compile_fullgraph (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.torch_compile_fullgraph", false]], "torch_compile_inductor_enabled (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.torch_compile_inductor_enabled", false]], "torch_compile_piecewise_cuda_graph (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.torch_compile_piecewise_cuda_graph", false]], "torchllmargs (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs", false]], "tp_split_dim() (tensorrt_llm.layers.linear.linear class method)": [[83, "tensorrt_llm.layers.linear.Linear.tp_split_dim", false]], "tp_split_dim() (tensorrt_llm.layers.linear.linearbase class method)": [[83, "tensorrt_llm.layers.linear.LinearBase.tp_split_dim", false]], "tp_split_dim() (tensorrt_llm.layers.linear.rowlinear class method)": [[83, "tensorrt_llm.layers.linear.RowLinear.tp_split_dim", false]], "transfer_mode (tensorrt_llm.llmapi.kvcacheretentionconfig property)": [[70, "tensorrt_llm.llmapi.KvCacheRetentionConfig.transfer_mode", false]], "transpose() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.transpose", false]], "transpose() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.transpose", false]], "trtllm-serve-disaggregated command line option": [[30, "cmdoption-trtllm-serve-disaggregated-c", false], [30, "cmdoption-trtllm-serve-disaggregated-r", false], [30, "cmdoption-trtllm-serve-disaggregated-t", false]], "trtllm-serve-disaggregated_mpi_worker command line option": [[30, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", false], [30, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", false]], "trtllm-serve-serve command line option": [[30, "cmdoption-trtllm-serve-serve-arg-MODEL", false], [30, "cmdoption-trtllm-serve-serve-backend", false], [30, "cmdoption-trtllm-serve-serve-cluster_size", false], [30, "cmdoption-trtllm-serve-serve-ep_size", false], [30, "cmdoption-trtllm-serve-serve-extra_llm_api_options", false], [30, "cmdoption-trtllm-serve-serve-gpus_per_node", false], [30, "cmdoption-trtllm-serve-serve-host", false], [30, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", false], [30, "cmdoption-trtllm-serve-serve-log_level", false], [30, "cmdoption-trtllm-serve-serve-max_batch_size", false], [30, "cmdoption-trtllm-serve-serve-max_beam_width", false], [30, "cmdoption-trtllm-serve-serve-max_num_tokens", false], [30, "cmdoption-trtllm-serve-serve-max_seq_len", false], [30, "cmdoption-trtllm-serve-serve-num_postprocess_workers", false], [30, "cmdoption-trtllm-serve-serve-port", false], [30, "cmdoption-trtllm-serve-serve-pp_size", false], [30, "cmdoption-trtllm-serve-serve-reasoning_parser", false], [30, "cmdoption-trtllm-serve-serve-tokenizer", false], [30, "cmdoption-trtllm-serve-serve-tp_size", false], [30, "cmdoption-trtllm-serve-serve-trust_remote_code", false]], "trtllm_modules_to_hf_modules (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.trtllm_modules_to_hf_modules", false]], "trtllmargs (class in tensorrt_llm.llmapi)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs", false]], "truncate_prompt_tokens (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.truncate_prompt_tokens", false]], "twoshot (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.TWOSHOT", false]], "ub (tensorrt_llm.functional.allreducestrategy attribute)": [[82, "tensorrt_llm.functional.AllReduceStrategy.UB", false]], "unary() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.unary", false]], "unbind() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.unbind", false]], "unbind() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.unbind", false]], "unfuse_qkv_projections() (tensorrt_llm.models.sd3transformer2dmodel method)": [[84, "tensorrt_llm.models.SD3Transformer2DModel.unfuse_qkv_projections", false]], "unpatchify() (tensorrt_llm.models.dit method)": [[84, "tensorrt_llm.models.DiT.unpatchify", false]], "unsqueeze() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.unsqueeze", false]], "unsqueeze() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.unsqueeze", false]], "update() (tensorrt_llm.llmapi.buildconfig method)": [[70, "tensorrt_llm.llmapi.BuildConfig.update", false]], "update() (tensorrt_llm.runtime.samplingconfig method)": [[87, "tensorrt_llm.runtime.SamplingConfig.update", false]], "update_from_dict() (tensorrt_llm.llmapi.buildconfig method)": [[70, "tensorrt_llm.llmapi.BuildConfig.update_from_dict", false]], "update_kv_cache_type() (tensorrt_llm.llmapi.buildconfig method)": [[70, "tensorrt_llm.llmapi.BuildConfig.update_kv_cache_type", false]], "update_output_ids_by_offset() (tensorrt_llm.runtime.generationsession method)": [[87, "tensorrt_llm.runtime.GenerationSession.update_output_ids_by_offset", false]], "update_strategy() (tensorrt_llm.functional.allreduceparams method)": [[82, "tensorrt_llm.functional.AllReduceParams.update_strategy", false]], "use_beam_hyps (tensorrt_llm.runtime.samplingconfig attribute)": [[87, "tensorrt_llm.runtime.SamplingConfig.use_beam_hyps", false]], "use_beam_search (tensorrt_llm.llmapi.samplingparams attribute)": [[70, "tensorrt_llm.llmapi.SamplingParams.use_beam_search", false]], "use_cuda_graph (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.use_cuda_graph", false]], "use_dynamic_tree (tensorrt_llm.llmapi.eagledecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.EagleDecodingConfig.use_dynamic_tree", false]], "use_gemm_allreduce_plugin (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.use_gemm_allreduce_plugin", false]], "use_gpt_attention_plugin (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.use_gpt_attention_plugin", false]], "use_kv_cache (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.use_kv_cache", false]], "use_kv_cache (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.use_kv_cache", false]], "use_lora() (tensorrt_llm.models.decodermodel method)": [[84, "tensorrt_llm.models.DecoderModel.use_lora", false]], "use_lora() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.use_lora", false]], "use_lora() (tensorrt_llm.models.gemmaforcausallm method)": [[84, "tensorrt_llm.models.GemmaForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.gptforcausallm method)": [[84, "tensorrt_llm.models.GPTForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.llamaforcausallm method)": [[84, "tensorrt_llm.models.LLaMAForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.mllamaforcausallm method)": [[84, "tensorrt_llm.models.MLLaMAForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.phi3forcausallm method)": [[84, "tensorrt_llm.models.Phi3ForCausalLM.use_lora", false]], "use_lora() (tensorrt_llm.models.phiforcausallm method)": [[84, "tensorrt_llm.models.PhiForCausalLM.use_lora", false]], "use_lora_plugin (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.use_lora_plugin", false]], "use_lora_plugin (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.use_lora_plugin", false]], "use_mamba_conv1d_plugin (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.use_mamba_conv1d_plugin", false]], "use_meta_recipe (tensorrt_llm.llmapi.quantconfig attribute)": [[70, "tensorrt_llm.llmapi.QuantConfig.use_meta_recipe", false]], "use_mrope (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.use_mrope", false]], "use_prompt_tuning() (tensorrt_llm.models.encodermodel method)": [[84, "tensorrt_llm.models.EncoderModel.use_prompt_tuning", false]], "use_refit (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.use_refit", false]], "use_relaxed_acceptance_for_thinking (tensorrt_llm.llmapi.mtpdecodingconfig attribute)": [[70, "tensorrt_llm.llmapi.MTPDecodingConfig.use_relaxed_acceptance_for_thinking", false]], "use_strip_plan (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.use_strip_plan", false]], "validate_cuda_graph_config() (tensorrt_llm.llmapi.torchllmargs method)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.validate_cuda_graph_config", false]], "validate_cuda_graph_max_batch_size() (tensorrt_llm.llmapi.torchllmargs class method)": [[70, "tensorrt_llm.llmapi.TorchLlmArgs.validate_cuda_graph_max_batch_size", false]], "validate_positive_values() (tensorrt_llm.llmapi.lookaheaddecodingconfig class method)": [[70, "tensorrt_llm.llmapi.LookaheadDecodingConfig.validate_positive_values", false]], "verbatim (tensorrt_llm.models.gemmaconfig attribute)": [[84, "tensorrt_llm.models.GemmaConfig.VERBATIM", false]], "video_preprocess() (tensorrt_llm.runtime.multimodalmodelrunner method)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.video_preprocess", false]], "view() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.view", false]], "view() (tensorrt_llm.functional.tensor method)": [[82, "tensorrt_llm.functional.Tensor.view", false]], "view() (tensorrt_llm.runtime.tensorinfo method)": [[87, "tensorrt_llm.runtime.TensorInfo.view", false]], "visual_engine_dir (tensorrt_llm.runtime.multimodalmodelrunner property)": [[87, "tensorrt_llm.runtime.MultimodalModelRunner.visual_engine_dir", false]], "visualize_network (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.visualize_network", false]], "vocab_size (tensorrt_llm.runtime.generationsession property)": [[87, "tensorrt_llm.runtime.GenerationSession.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelconfig attribute)": [[87, "tensorrt_llm.runtime.ModelConfig.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.vocab_size", false]], "vocab_size (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.vocab_size", false]], "vocab_size_padded (tensorrt_llm.runtime.modelrunner property)": [[87, "tensorrt_llm.runtime.ModelRunner.vocab_size_padded", false]], "vocab_size_padded (tensorrt_llm.runtime.modelrunnercpp property)": [[87, "tensorrt_llm.runtime.ModelRunnerCpp.vocab_size_padded", false]], "w4a16 (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A16", false]], "w4a16_awq (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A16_AWQ", false]], "w4a16_gptq (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A16_GPTQ", false]], "w4a8_awq (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A8_AWQ", false]], "w4a8_qserve_per_channel (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A8_QSERVE_PER_CHANNEL", false]], "w4a8_qserve_per_group (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W4A8_QSERVE_PER_GROUP", false]], "w8a16 (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A16", false]], "w8a16_gptq (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A16_GPTQ", false]], "w8a8_sq_per_channel (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL", false]], "w8a8_sq_per_channel_per_tensor_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL_PER_TENSOR_PLUGIN", false]], "w8a8_sq_per_channel_per_token_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_CHANNEL_PER_TOKEN_PLUGIN", false]], "w8a8_sq_per_tensor_per_token_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_TENSOR_PER_TOKEN_PLUGIN", false]], "w8a8_sq_per_tensor_plugin (tensorrt_llm.llmapi.quantalgo attribute)": [[70, "tensorrt_llm.llmapi.QuantAlgo.W8A8_SQ_PER_TENSOR_PLUGIN", false]], "weight_loader() (tensorrt_llm.layers.attention.deepseekv2attention method)": [[83, "tensorrt_llm.layers.attention.DeepseekV2Attention.weight_loader", false]], "weight_loader() (tensorrt_llm.layers.embedding.embedding method)": [[83, "tensorrt_llm.layers.embedding.Embedding.weight_loader", false]], "weight_loader() (tensorrt_llm.layers.linear.linearbase method)": [[83, "tensorrt_llm.layers.linear.LinearBase.weight_loader", false]], "weight_sparsity (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.weight_sparsity", false]], "weight_streaming (tensorrt_llm.llmapi.buildconfig attribute)": [[70, "tensorrt_llm.llmapi.BuildConfig.weight_streaming", false]], "where() (in module tensorrt_llm.functional)": [[82, "tensorrt_llm.functional.where", false]], "whisperencoder (class in tensorrt_llm.models)": [[84, "tensorrt_llm.models.WhisperEncoder", false]], "workspace (tensorrt_llm.llmapi.llm attribute)": [[70, "tensorrt_llm.llmapi.LLM.workspace", false]], "workspace (tensorrt_llm.llmapi.llm property)": [[70, "id1", false]], "workspace (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "tensorrt_llm.llmapi.TrtLlmArgs.workspace", false]], "wrapped_property (tensorrt_llm.llmapi.torchllmargs attribute)": [[70, "id11", false], [70, "id14", false], [70, "id17", false], [70, "tensorrt_llm.llmapi.TorchLlmArgs.wrapped_property", false]], "wrapped_property (tensorrt_llm.llmapi.trtllmargs attribute)": [[70, "id20", false], [70, "id23", false], [70, "id26", false], [70, "id29", false], [70, "id32", false], [70, "tensorrt_llm.llmapi.TrtLlmArgs.wrapped_property", false]], "yarn (tensorrt_llm.functional.positionembeddingtype attribute)": [[82, "tensorrt_llm.functional.PositionEmbeddingType.yarn", false]], "yarn (tensorrt_llm.functional.rotaryscalingtype attribute)": [[82, "tensorrt_llm.functional.RotaryScalingType.yarn", false]]}, "objects": {"": [[1, 0, 1, "c.FMT_DIM", "FMT_DIM"], [1, 0, 1, "c.SET_FROM_OPTIONAL", "SET_FROM_OPTIONAL"], [1, 1, 1, "_CPPv48nvinfer1", "nvinfer1"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [1, 1, 1, "_CPPv412tensorrt_llm", "tensorrt_llm"], [0, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [1, 1, 1, "_CPPv4N12tensorrt_llm13batch_managerE", "tensorrt_llm::batch_manager"], [0, 1, 1, "_CPPv4N12tensorrt_llm13batch_manager16kv_cache_managerE", "tensorrt_llm::batch_manager::kv_cache_manager"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executorE", "tensorrt_llm::executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutputE", "tensorrt_llm::executor::AdditionalModelOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput::gatherContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput21AdditionalModelOutputENSt6stringEb", "tensorrt_llm::executor::AdditionalModelOutput::AdditionalModelOutput::name"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput13gatherContextE", "tensorrt_llm::executor::AdditionalModelOutput::gatherContext"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21AdditionalModelOutput4nameE", "tensorrt_llm::executor::AdditionalModelOutput::name"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", "tensorrt_llm::executor::AdditionalModelOutput::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor21AdditionalModelOutputeqERK21AdditionalModelOutput", "tensorrt_llm::executor::AdditionalModelOutput::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputE", "tensorrt_llm::executor::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::name"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput16AdditionalOutputENSt6stringE6Tensor", "tensorrt_llm::executor::AdditionalOutput::AdditionalOutput::output"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput4nameE", "tensorrt_llm::executor::AdditionalOutput::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERK16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputaSERR16AdditionalOutput", "tensorrt_llm::executor::AdditionalOutput::operator=::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutput6outputE", "tensorrt_llm::executor::AdditionalOutput::output"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor16AdditionalOutputD0Ev", "tensorrt_llm::executor::AdditionalOutput::~AdditionalOutput"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12BatchingTypeE", "tensorrt_llm::executor::BatchingType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12BatchingType9kINFLIGHTE", "tensorrt_llm::executor::BatchingType::kINFLIGHT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12BatchingType7kSTATICE", "tensorrt_llm::executor::BatchingType::kSTATIC"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10BeamTokensE", "tensorrt_llm::executor::BeamTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10BufferViewE", "tensorrt_llm::executor::BufferView"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfigE", "tensorrt_llm::executor::CacheTransceiverConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", "tensorrt_llm::executor::CacheTransceiverConfig::CacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig22CacheTransceiverConfigENSt8optionalI6size_tEE", "tensorrt_llm::executor::CacheTransceiverConfig::CacheTransceiverConfig::maxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfig15getMaxNumTokensEv", "tensorrt_llm::executor::CacheTransceiverConfig::getMaxNumTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig13mMaxNumTokensE", "tensorrt_llm::executor::CacheTransceiverConfig::mMaxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", "tensorrt_llm::executor::CacheTransceiverConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22CacheTransceiverConfigeqERK22CacheTransceiverConfig", "tensorrt_llm::executor::CacheTransceiverConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", "tensorrt_llm::executor::CacheTransceiverConfig::setMaxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22CacheTransceiverConfig15setMaxNumTokensE6size_t", "tensorrt_llm::executor::CacheTransceiverConfig::setMaxNumTokens::maxNumTokens"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicyE", "tensorrt_llm::executor::CapacitySchedulerPolicy"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy20kGUARANTEED_NO_EVICTE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kGUARANTEED_NO_EVICT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy16kMAX_UTILIZATIONE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kMAX_UTILIZATION"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor23CapacitySchedulerPolicy13kSTATIC_BATCHE", "tensorrt_llm::executor::CapacitySchedulerPolicy::kSTATIC_BATCH"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationModeE", "tensorrt_llm::executor::CommunicationMode"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationMode7kLEADERE", "tensorrt_llm::executor::CommunicationMode::kLEADER"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationMode13kORCHESTRATORE", "tensorrt_llm::executor::CommunicationMode::kORCHESTRATOR"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationTypeE", "tensorrt_llm::executor::CommunicationType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor17CommunicationType4kMPIE", "tensorrt_llm::executor::CommunicationType::kMPI"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicyE", "tensorrt_llm::executor::ContextChunkingPolicy"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy15kEQUAL_PROGRESSE", "tensorrt_llm::executor::ContextChunkingPolicy::kEQUAL_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor21ContextChunkingPolicy24kFIRST_COME_FIRST_SERVEDE", "tensorrt_llm::executor::ContextChunkingPolicy::kFIRST_COME_FIRST_SERVED"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsE", "tensorrt_llm::executor::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsERR18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::draftTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::firstGenTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::reqId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypeRKNSt6vectorIcEENSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::serializedState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams18ContextPhaseParamsE9VecTokens13RequestIdTypePvNSt8optionalI9VecTokensEE", "tensorrt_llm::executor::ContextPhaseParams::ContextPhaseParams::state"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams13RequestIdTypeE", "tensorrt_llm::executor::ContextPhaseParams::RequestIdType"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8StatePtrE", "tensorrt_llm::executor::ContextPhaseParams::StatePtr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", "tensorrt_llm::executor::ContextPhaseParams::deleter"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams7deleterEPKv", "tensorrt_llm::executor::ContextPhaseParams::deleter::data"], [0, 3, 1, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams14getDraftTokensEv", "tensorrt_llm::executor::ContextPhaseParams::getDraftTokens"], [0, 3, 1, "_CPPv4NKR12tensorrt_llm8executor18ContextPhaseParams17getFirstGenTokensEv", "tensorrt_llm::executor::ContextPhaseParams::getFirstGenTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getReqIdEv", "tensorrt_llm::executor::ContextPhaseParams::getReqId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams18getSerializedStateEv", "tensorrt_llm::executor::ContextPhaseParams::getSerializedState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams8getStateEv", "tensorrt_llm::executor::ContextPhaseParams::getState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParams8getStateEv", "tensorrt_llm::executor::ContextPhaseParams::getState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12mDraftTokensE", "tensorrt_llm::executor::ContextPhaseParams::mDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams15mFirstGenTokensE", "tensorrt_llm::executor::ContextPhaseParams::mFirstGenTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mReqIdE", "tensorrt_llm::executor::ContextPhaseParams::mReqId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams6mStateE", "tensorrt_llm::executor::ContextPhaseParams::mState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsaSERR18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator="], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18ContextPhaseParamseqERK18ContextPhaseParams", "tensorrt_llm::executor::ContextPhaseParams::operator=="], [0, 3, 1, "_CPPv4NO12tensorrt_llm8executor18ContextPhaseParams17popFirstGenTokensEv", "tensorrt_llm::executor::ContextPhaseParams::popFirstGenTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParams12releaseStateEv", "tensorrt_llm::executor::ContextPhaseParams::releaseState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18ContextPhaseParamsD0Ev", "tensorrt_llm::executor::ContextPhaseParams::~ContextPhaseParams"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverStateE", "tensorrt_llm::executor::DataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEv", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState::cacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState20DataTransceiverStateEN8kv_cache10CacheStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::DataTransceiverState::commState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState13getCacheStateEv", "tensorrt_llm::executor::DataTransceiverState::getCacheState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState12getCommStateEv", "tensorrt_llm::executor::DataTransceiverState::getCommState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState11mCacheStateE", "tensorrt_llm::executor::DataTransceiverState::mCacheState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState10mCommStateE", "tensorrt_llm::executor::DataTransceiverState::mCommState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", "tensorrt_llm::executor::DataTransceiverState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverStateeqERK20DataTransceiverState", "tensorrt_llm::executor::DataTransceiverState::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", "tensorrt_llm::executor::DataTransceiverState::setCacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState13setCacheStateEN8kv_cache10CacheStateE", "tensorrt_llm::executor::DataTransceiverState::setCacheState::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::setCommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20DataTransceiverState12setCommStateEN8kv_cache9CommStateE", "tensorrt_llm::executor::DataTransceiverState::setCommState::state"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20DataTransceiverState8toStringEv", "tensorrt_llm::executor::DataTransceiverState::toString"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8DataTypeE", "tensorrt_llm::executor::DataType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kBF16E", "tensorrt_llm::executor::DataType::kBF16"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kBOOLE", "tensorrt_llm::executor::DataType::kBOOL"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kFP16E", "tensorrt_llm::executor::DataType::kFP16"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kFP32E", "tensorrt_llm::executor::DataType::kFP32"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType4kFP8E", "tensorrt_llm::executor::DataType::kFP8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kINT32E", "tensorrt_llm::executor::DataType::kINT32"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kINT64E", "tensorrt_llm::executor::DataType::kINT64"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType5kINT8E", "tensorrt_llm::executor::DataType::kINT8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType6kUINT8E", "tensorrt_llm::executor::DataType::kUINT8"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8DataType8kUNKNOWNE", "tensorrt_llm::executor::DataType::kUNKNOWN"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfigE", "tensorrt_llm::executor::DebugConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugInputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugOutputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugTensorNames"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig11DebugConfigEbb9StringVec10SizeType32", "tensorrt_llm::executor::DebugConfig::DebugConfig::debugTensorsMaxIterations"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig9StringVecE", "tensorrt_llm::executor::DebugConfig::StringVec"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig20getDebugInputTensorsEv", "tensorrt_llm::executor::DebugConfig::getDebugInputTensors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig21getDebugOutputTensorsEv", "tensorrt_llm::executor::DebugConfig::getDebugOutputTensors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig19getDebugTensorNamesEv", "tensorrt_llm::executor::DebugConfig::getDebugTensorNames"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfig28getDebugTensorsMaxIterationsEv", "tensorrt_llm::executor::DebugConfig::getDebugTensorsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig18mDebugInputTensorsE", "tensorrt_llm::executor::DebugConfig::mDebugInputTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19mDebugOutputTensorsE", "tensorrt_llm::executor::DebugConfig::mDebugOutputTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig17mDebugTensorNamesE", "tensorrt_llm::executor::DebugConfig::mDebugTensorNames"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig26mDebugTensorsMaxIterationsE", "tensorrt_llm::executor::DebugConfig::mDebugTensorsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", "tensorrt_llm::executor::DebugConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor11DebugConfigeqERK11DebugConfig", "tensorrt_llm::executor::DebugConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugInputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig20setDebugInputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugInputTensors::debugInputTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugOutputTensors"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig21setDebugOutputTensorsEb", "tensorrt_llm::executor::DebugConfig::setDebugOutputTensors::debugOutputTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", "tensorrt_llm::executor::DebugConfig::setDebugTensorNames"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig19setDebugTensorNamesERK9StringVec", "tensorrt_llm::executor::DebugConfig::setDebugTensorNames::debugTensorNames"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", "tensorrt_llm::executor::DebugConfig::setDebugTensorsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11DebugConfig28setDebugTensorsMaxIterationsE10SizeType32", "tensorrt_llm::executor::DebugConfig::setDebugTensorsMaxIterations::debugTensorsMaxIterations"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIterationE", "tensorrt_llm::executor::DebugTensorsPerIteration"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration12debugTensorsE", "tensorrt_llm::executor::DebugTensorsPerIteration::debugTensors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24DebugTensorsPerIteration4iterE", "tensorrt_llm::executor::DebugTensorsPerIteration::iter"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfigE", "tensorrt_llm::executor::DecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14DecodingConfigENSt8optionalI12DecodingModeEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI13MedusaChoicesEENSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::DecodingConfig::DecodingConfig::medusaChoices"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31enableSeamlessLookaheadDecodingEv", "tensorrt_llm::executor::DecodingConfig::enableSeamlessLookaheadDecoding"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig15getDecodingModeEv", "tensorrt_llm::executor::DecodingConfig::getDecodingMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig14getEagleConfigEv", "tensorrt_llm::executor::DecodingConfig::getEagleConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig26getLookaheadDecodingConfigEv", "tensorrt_llm::executor::DecodingConfig::getLookaheadDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig33getLookaheadDecodingMaxNumRequestEv", "tensorrt_llm::executor::DecodingConfig::getLookaheadDecodingMaxNumRequest"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfig16getMedusaChoicesEv", "tensorrt_llm::executor::DecodingConfig::getMedusaChoices"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig13mDecodingModeE", "tensorrt_llm::executor::DecodingConfig::mDecodingMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig12mEagleConfigE", "tensorrt_llm::executor::DecodingConfig::mEagleConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig24mLookaheadDecodingConfigE", "tensorrt_llm::executor::DecodingConfig::mLookaheadDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig31mLookaheadDecodingMaxNumRequestE", "tensorrt_llm::executor::DecodingConfig::mLookaheadDecodingMaxNumRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14mMedusaChoicesE", "tensorrt_llm::executor::DecodingConfig::mMedusaChoices"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", "tensorrt_llm::executor::DecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor14DecodingConfigeqERK14DecodingConfig", "tensorrt_llm::executor::DecodingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig15setDecodingModeERK12DecodingMode", "tensorrt_llm::executor::DecodingConfig::setDecodingMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig14setEagleConfigERK11EagleConfig", "tensorrt_llm::executor::DecodingConfig::setEagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::DecodingConfig::setLookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig26setLookaheadDecodingConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::DecodingConfig::setLookaheadDecodingConfig::lookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14DecodingConfig16setMedusaChoicesERK13MedusaChoices", "tensorrt_llm::executor::DecodingConfig::setMedusaChoices"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12DecodingModeE", "tensorrt_llm::executor::DecodingMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4AutoEv", "tensorrt_llm::executor::DecodingMode::Auto"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode10BeamSearchEv", "tensorrt_llm::executor::DecodingMode::BeamSearch"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::DecodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12DecodingModeE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::DecodingMode::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5EagleEv", "tensorrt_llm::executor::DecodingMode::Eagle"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExplicitDraftTokensEv", "tensorrt_llm::executor::DecodingMode::ExplicitDraftTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19ExternalDraftTokensEv", "tensorrt_llm::executor::DecodingMode::ExternalDraftTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9LookaheadEv", "tensorrt_llm::executor::DecodingMode::Lookahead"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6MedusaEv", "tensorrt_llm::executor::DecodingMode::Medusa"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopKEv", "tensorrt_llm::executor::DecodingMode::TopK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8TopKTopPEv", "tensorrt_llm::executor::DecodingMode::TopKTopP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode4TopPEv", "tensorrt_llm::executor::DecodingMode::TopP"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14UnderlyingTypeE", "tensorrt_llm::executor::DecodingMode::UnderlyingType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::allBitSet"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::allBitSet::bits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::anyBitSet"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::executor::DecodingMode::anyBitSet::bits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7getNameEv", "tensorrt_llm::executor::DecodingMode::getName"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8getStateEv", "tensorrt_llm::executor::DecodingMode::getState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isAutoEv", "tensorrt_llm::executor::DecodingMode::isAuto"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isBeamSearchEv", "tensorrt_llm::executor::DecodingMode::isBeamSearch"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode7isEagleEv", "tensorrt_llm::executor::DecodingMode::isEagle"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExplicitDraftTokensEv", "tensorrt_llm::executor::DecodingMode::isExplicitDraftTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isExternalDraftTokensEv", "tensorrt_llm::executor::DecodingMode::isExternalDraftTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode11isLookaheadEv", "tensorrt_llm::executor::DecodingMode::isLookahead"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode8isMedusaEv", "tensorrt_llm::executor::DecodingMode::isMedusa"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopKEv", "tensorrt_llm::executor::DecodingMode::isTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isTopKandTopPEv", "tensorrt_llm::executor::DecodingMode::isTopKandTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isTopKorTopPEv", "tensorrt_llm::executor::DecodingMode::isTopKorTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode6isTopPEv", "tensorrt_llm::executor::DecodingMode::isTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseBanTokensEv", "tensorrt_llm::executor::DecodingMode::isUseBanTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode13isUseBanWordsEv", "tensorrt_llm::executor::DecodingMode::isUseBanWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUseExplicitEosStopEv", "tensorrt_llm::executor::DecodingMode::isUseExplicitEosStop"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode21isUseFrequencyPenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseFrequencyPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode18isUseMaxLengthStopEv", "tensorrt_llm::executor::DecodingMode::isUseMaxLengthStop"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseMinLengthEv", "tensorrt_llm::executor::DecodingMode::isUseMinLength"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode9isUseMinPEv", "tensorrt_llm::executor::DecodingMode::isUseMinP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseNoRepeatNgramSizeEv", "tensorrt_llm::executor::DecodingMode::isUseNoRepeatNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseOccurrencePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseOccurrencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode12isUsePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUsePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode20isUsePresencePenaltyEv", "tensorrt_llm::executor::DecodingMode::isUsePresencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode22isUseRepetitionPenaltyEv", "tensorrt_llm::executor::DecodingMode::isUseRepetitionPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode17isUseStopCriteriaEv", "tensorrt_llm::executor::DecodingMode::isUseStopCriteria"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode14isUseStopWordsEv", "tensorrt_llm::executor::DecodingMode::isUseStopWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode16isUseTemperatureEv", "tensorrt_llm::executor::DecodingMode::isUseTemperature"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingMode28isUseVariableBeamWidthSearchEv", "tensorrt_llm::executor::DecodingMode::isUseVariableBeamWidthSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kAutoE", "tensorrt_llm::executor::DecodingMode::kAuto"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11kBeamSearchE", "tensorrt_llm::executor::DecodingMode::kBeamSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6kEagleE", "tensorrt_llm::executor::DecodingMode::kEagle"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExplicitDraftTokensE", "tensorrt_llm::executor::DecodingMode::kExplicitDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20kExternalDraftTokensE", "tensorrt_llm::executor::DecodingMode::kExternalDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode10kLookaheadE", "tensorrt_llm::executor::DecodingMode::kLookahead"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7kMedusaE", "tensorrt_llm::executor::DecodingMode::kMedusa"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kNumFlagsE", "tensorrt_llm::executor::DecodingMode::kNumFlags"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopKE", "tensorrt_llm::executor::DecodingMode::kTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode9kTopKTopPE", "tensorrt_llm::executor::DecodingMode::kTopKTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode5kTopPE", "tensorrt_llm::executor::DecodingMode::kTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseBanTokensE", "tensorrt_llm::executor::DecodingMode::kUseBanTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12kUseBanWordsE", "tensorrt_llm::executor::DecodingMode::kUseBanWords"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19kUseExplicitEosStopE", "tensorrt_llm::executor::DecodingMode::kUseExplicitEosStop"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22kUseFrequencyPenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseFrequencyPenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode17kUseMaxLengthStopE", "tensorrt_llm::executor::DecodingMode::kUseMaxLengthStop"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseMinLengthE", "tensorrt_llm::executor::DecodingMode::kUseMinLength"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8kUseMinPE", "tensorrt_llm::executor::DecodingMode::kUseMinP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUseNoRepeatNgramSizeE", "tensorrt_llm::executor::DecodingMode::kUseNoRepeatNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseOccurrencePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseOccurrencePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUsePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUsePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode21kUsePresencePenaltiesE", "tensorrt_llm::executor::DecodingMode::kUsePresencePenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode23kUseRepetitionPenaltiesE", "tensorrt_llm::executor::DecodingMode::kUseRepetitionPenalties"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode24kUseStandardStopCriteriaE", "tensorrt_llm::executor::DecodingMode::kUseStandardStopCriteria"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode13kUseStopWordsE", "tensorrt_llm::executor::DecodingMode::kUseStopWords"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode15kUseTemperatureE", "tensorrt_llm::executor::DecodingMode::kUseTemperature"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode27kUseVariableBeamWidthSearchE", "tensorrt_llm::executor::DecodingMode::kUseVariableBeamWidthSearch"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode6mStateE", "tensorrt_llm::executor::DecodingMode::mState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", "tensorrt_llm::executor::DecodingMode::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor12DecodingModeeqERK12DecodingMode", "tensorrt_llm::executor::DecodingMode::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode8setBitToE14UnderlyingTypeb", "tensorrt_llm::executor::DecodingMode::setBitTo::x"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", "tensorrt_llm::executor::DecodingMode::useBanTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useBanTokensEb", "tensorrt_llm::executor::DecodingMode::useBanTokens::banTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", "tensorrt_llm::executor::DecodingMode::useBanWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode11useBanWordsEb", "tensorrt_llm::executor::DecodingMode::useBanWords::banWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", "tensorrt_llm::executor::DecodingMode::useExplicitEosStop"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18useExplicitEosStopEb", "tensorrt_llm::executor::DecodingMode::useExplicitEosStop::explicitEosStop"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", "tensorrt_llm::executor::DecodingMode::useFrequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode19useFrequencyPenaltyEb", "tensorrt_llm::executor::DecodingMode::useFrequencyPenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", "tensorrt_llm::executor::DecodingMode::useMaxLengthStop"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode16useMaxLengthStopEb", "tensorrt_llm::executor::DecodingMode::useMaxLengthStop::maxLengthStop"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", "tensorrt_llm::executor::DecodingMode::useMinLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useMinLengthEb", "tensorrt_llm::executor::DecodingMode::useMinLength::useMinLen"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", "tensorrt_llm::executor::DecodingMode::useMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode7useMinPEb", "tensorrt_llm::executor::DecodingMode::useMinP::useMinP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", "tensorrt_llm::executor::DecodingMode::useNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useNoRepeatNgramSizeEb", "tensorrt_llm::executor::DecodingMode::useNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", "tensorrt_llm::executor::DecodingMode::useOccurrencePenalties"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode22useOccurrencePenaltiesEb", "tensorrt_llm::executor::DecodingMode::useOccurrencePenalties::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", "tensorrt_llm::executor::DecodingMode::usePresencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode18usePresencePenaltyEb", "tensorrt_llm::executor::DecodingMode::usePresencePenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", "tensorrt_llm::executor::DecodingMode::useRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode20useRepetitionPenaltyEb", "tensorrt_llm::executor::DecodingMode::useRepetitionPenalty::usePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", "tensorrt_llm::executor::DecodingMode::useStopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode12useStopWordsEb", "tensorrt_llm::executor::DecodingMode::useStopWords::stopWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", "tensorrt_llm::executor::DecodingMode::useTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode14useTemperatureEb", "tensorrt_llm::executor::DecodingMode::useTemperature::useTemp"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", "tensorrt_llm::executor::DecodingMode::useVariableBeamWidthSearch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12DecodingMode26useVariableBeamWidthSearchEb", "tensorrt_llm::executor::DecodingMode::useVariableBeamWidthSearch::useVariableBeamWidthSearch"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStatsE", "tensorrt_llm::executor::DisServingRequestStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats11kvCacheSizeE", "tensorrt_llm::executor::DisServingRequestStats::kvCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22DisServingRequestStats17kvCacheTransferMSE", "tensorrt_llm::executor::DisServingRequestStats::kvCacheTransferMS"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfigE", "tensorrt_llm::executor::DynamicBatchConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::batchSizeTable"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::dynamicBatchMovingAverageWindow"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::enableBatchSizeTuning"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig18DynamicBatchConfigEbb10SizeType32NSt6vectorINSt4pairI10SizeType3210SizeType32EEEE", "tensorrt_llm::executor::DynamicBatchConfig::DynamicBatchConfig::enableMaxNumTokensTuning"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig17getBatchSizeTableEv", "tensorrt_llm::executor::DynamicBatchConfig::getBatchSizeTable"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig34getDynamicBatchMovingAverageWindowEv", "tensorrt_llm::executor::DynamicBatchConfig::getDynamicBatchMovingAverageWindow"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig24getEnableBatchSizeTuningEv", "tensorrt_llm::executor::DynamicBatchConfig::getEnableBatchSizeTuning"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18DynamicBatchConfig27getEnableMaxNumTokensTuningEv", "tensorrt_llm::executor::DynamicBatchConfig::getEnableMaxNumTokensTuning"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22kDefaultBatchSizeTableE", "tensorrt_llm::executor::DynamicBatchConfig::kDefaultBatchSizeTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig39kDefaultDynamicBatchMovingAverageWindowE", "tensorrt_llm::executor::DynamicBatchConfig::kDefaultDynamicBatchMovingAverageWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig15mBatchSizeTableE", "tensorrt_llm::executor::DynamicBatchConfig::mBatchSizeTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig32mDynamicBatchMovingAverageWindowE", "tensorrt_llm::executor::DynamicBatchConfig::mDynamicBatchMovingAverageWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig22mEnableBatchSizeTuningE", "tensorrt_llm::executor::DynamicBatchConfig::mEnableBatchSizeTuning"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18DynamicBatchConfig25mEnableMaxNumTokensTuningE", "tensorrt_llm::executor::DynamicBatchConfig::mEnableMaxNumTokensTuning"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12EagleChoicesE", "tensorrt_llm::executor::EagleChoices"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfigE", "tensorrt_llm::executor::EagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::dynamicTreeMaxTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::eagleChoices"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::greedySampling"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::posteriorThreshold"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig11EagleConfigENSt8optionalI12EagleChoicesEEbNSt8optionalIfEEbNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::EagleConfig::EagleConfig::useDynamicTree"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", "tensorrt_llm::executor::EagleConfig::checkPosteriorValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19checkPosteriorValueERKNSt8optionalIfEE", "tensorrt_llm::executor::EagleConfig::checkPosteriorValue::value"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getDynamicTreeMaxTopKEv", "tensorrt_llm::executor::EagleConfig::getDynamicTreeMaxTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig15getEagleChoicesEv", "tensorrt_llm::executor::EagleConfig::getEagleChoices"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig21getPosteriorThresholdEv", "tensorrt_llm::executor::EagleConfig::getPosteriorThreshold"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig16isGreedySamplingEv", "tensorrt_llm::executor::EagleConfig::isGreedySampling"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mDynamicTreeMaxTopKE", "tensorrt_llm::executor::EagleConfig::mDynamicTreeMaxTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig13mEagleChoicesE", "tensorrt_llm::executor::EagleConfig::mEagleChoices"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mGreedySamplingE", "tensorrt_llm::executor::EagleConfig::mGreedySampling"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig19mPosteriorThresholdE", "tensorrt_llm::executor::EagleConfig::mPosteriorThreshold"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11EagleConfig15mUseDynamicTreeE", "tensorrt_llm::executor::EagleConfig::mUseDynamicTree"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", "tensorrt_llm::executor::EagleConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfigeqERK11EagleConfig", "tensorrt_llm::executor::EagleConfig::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11EagleConfig14useDynamicTreeEv", "tensorrt_llm::executor::EagleConfig::useDynamicTree"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8ExecutorE", "tensorrt_llm::executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERR8Executor", "tensorrt_llm::executor::Executor::Executor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderEngineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderJsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderModel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::decoderModelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderEngineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderJsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderModel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::encoderModelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::engineBuffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK8Executor", "tensorrt_llm::executor::Executor::Executor::executor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::jsonConfigStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::managedWeights"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorENSt10shared_ptrI5ModelEERK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::model"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelPath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfigRKNSt8optionalINSt3mapINSt6stringE6TensorEEEE", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERK10BufferViewRKNSt6stringERK10BufferViewRKNSt6stringE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor8ExecutorERKNSt10filesystem4pathERKNSt10filesystem4pathE9ModelTypeRK14ExecutorConfig", "tensorrt_llm::executor::Executor::Executor::modelType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::requestId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::requestIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERK6IdTypeRKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt6vectorI6IdTypeEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14awaitResponsesERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::Executor::awaitResponses::timeout"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor18canEnqueueRequestsEv", "tensorrt_llm::executor::Executor::canEnqueueRequests"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", "tensorrt_llm::executor::Executor::cancelRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor13cancelRequestE6IdType", "tensorrt_llm::executor::Executor::cancelRequest::requestId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", "tensorrt_llm::executor::Executor::enqueueRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor14enqueueRequestERK7Request", "tensorrt_llm::executor::Executor::enqueueRequest::request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", "tensorrt_llm::executor::Executor::enqueueRequests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Executor15enqueueRequestsERKNSt6vectorI7RequestEE", "tensorrt_llm::executor::Executor::enqueueRequests::requests"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor22getKVCacheEventManagerEv", "tensorrt_llm::executor::Executor::getKVCacheEventManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestDebugTensorsEv", "tensorrt_llm::executor::Executor::getLatestDebugTensors"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor23getLatestIterationStatsEv", "tensorrt_llm::executor::Executor::getLatestIterationStats"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor21getLatestRequestStatsEv", "tensorrt_llm::executor::Executor::getLatestRequestStats"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Executor::getNumResponsesReady"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8Executor20getNumResponsesReadyERKNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Executor::getNumResponsesReady::requestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Executor13isParticipantEv", "tensorrt_llm::executor::Executor::isParticipant"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8Executor5mImplE", "tensorrt_llm::executor::Executor::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", "tensorrt_llm::executor::Executor::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERR8Executor", "tensorrt_llm::executor::Executor::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ExecutoraSERK8Executor", "tensorrt_llm::executor::Executor::operator=::executor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Executor8shutdownEv", "tensorrt_llm::executor::Executor::shutdown"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ExecutorD0Ev", "tensorrt_llm::executor::Executor::~Executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfigE", "tensorrt_llm::executor::ExecutorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::additionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::batchingType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::enableChunkedContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::enableTrtOverlap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::gatherGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::gpuWeightsPercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::iterStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::logitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxBatchSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxQueueSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::maxSeqIdleMicroseconds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::normalizeLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::promptTableOffloading"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::recvPollPeriodMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::requestStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14ExecutorConfigE10SizeType3215SchedulerConfig13KvCacheConfigbb10SizeType3210SizeType3212BatchingTypeNSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI14ParallelConfigEERKNSt8optionalI15PeftCacheConfigEENSt8optionalI25LogitsPostProcessorConfigEENSt8optionalI14DecodingConfigEEbfNSt8optionalI10SizeType32EERK29ExtendedRuntimePerfKnobConfigNSt8optionalI11DebugConfigEE10SizeType328uint64_tNSt8optionalI25SpeculativeDecodingConfigEENSt8optionalI20GuidedDecodingConfigEENSt8optionalINSt6vectorI21AdditionalModelOutputEEEENSt8optionalI22CacheTransceiverConfigEEbbb", "tensorrt_llm::executor::ExecutorConfig::ExecutorConfig::useGpuDirectStorage"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getAdditionalModelOutputsEv", "tensorrt_llm::executor::ExecutorConfig::getAdditionalModelOutputs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getBatchingTypeEv", "tensorrt_llm::executor::ExecutorConfig::getBatchingType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getCacheTransceiverConfigEv", "tensorrt_llm::executor::ExecutorConfig::getCacheTransceiverConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig14getDebugConfigEv", "tensorrt_llm::executor::ExecutorConfig::getDebugConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getDecodingConfigEv", "tensorrt_llm::executor::ExecutorConfig::getDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getEnableChunkedContextEv", "tensorrt_llm::executor::ExecutorConfig::getEnableChunkedContext"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getEnableTrtOverlapEv", "tensorrt_llm::executor::ExecutorConfig::getEnableTrtOverlap"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig32getExtendedRuntimePerfKnobConfigEv", "tensorrt_llm::executor::ExecutorConfig::getExtendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getGatherGenerationLogitsEv", "tensorrt_llm::executor::ExecutorConfig::getGatherGenerationLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getGpuWeightsPercentEv", "tensorrt_llm::executor::ExecutorConfig::getGpuWeightsPercent"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig23getGuidedDecodingConfigEv", "tensorrt_llm::executor::ExecutorConfig::getGuidedDecodingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getIterStatsMaxIterationsEv", "tensorrt_llm::executor::ExecutorConfig::getIterStatsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getKvCacheConfigEv", "tensorrt_llm::executor::ExecutorConfig::getKvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19getKvCacheConfigRefEv", "tensorrt_llm::executor::ExecutorConfig::getKvCacheConfigRef"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getLogitsPostProcessorConfigEv", "tensorrt_llm::executor::ExecutorConfig::getLogitsPostProcessorConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBatchSizeEv", "tensorrt_llm::executor::ExecutorConfig::getMaxBatchSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxBeamWidthEv", "tensorrt_llm::executor::ExecutorConfig::getMaxBeamWidth"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxNumTokensEv", "tensorrt_llm::executor::ExecutorConfig::getMaxNumTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig15getMaxQueueSizeEv", "tensorrt_llm::executor::ExecutorConfig::getMaxQueueSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig25getMaxSeqIdleMicrosecondsEv", "tensorrt_llm::executor::ExecutorConfig::getMaxSeqIdleMicroseconds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig20getNormalizeLogProbsEv", "tensorrt_llm::executor::ExecutorConfig::getNormalizeLogProbs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig17getParallelConfigEv", "tensorrt_llm::executor::ExecutorConfig::getParallelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getPeftCacheConfigEv", "tensorrt_llm::executor::ExecutorConfig::getPeftCacheConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig24getPromptTableOffloadingEv", "tensorrt_llm::executor::ExecutorConfig::getPromptTableOffloading"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig19getRecvPollPeriodMsEv", "tensorrt_llm::executor::ExecutorConfig::getRecvPollPeriodMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig28getRequestStatsMaxIterationsEv", "tensorrt_llm::executor::ExecutorConfig::getRequestStatsMaxIterations"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig18getSchedulerConfigEv", "tensorrt_llm::executor::ExecutorConfig::getSchedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21getSchedulerConfigRefEv", "tensorrt_llm::executor::ExecutorConfig::getSchedulerConfigRef"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig16getSpecDecConfigEv", "tensorrt_llm::executor::ExecutorConfig::getSpecDecConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ExecutorConfig22getUseGpuDirectStorageEv", "tensorrt_llm::executor::ExecutorConfig::getUseGpuDirectStorage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultIterStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultIterStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30kDefaultMaxSeqIdleMicrosecondsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultMaxSeqIdleMicroseconds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig33kDefaultRequestStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::kDefaultRequestStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mAdditionalModelOutputsE", "tensorrt_llm::executor::ExecutorConfig::mAdditionalModelOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mBatchingTypeE", "tensorrt_llm::executor::ExecutorConfig::mBatchingType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mCacheTransceiverConfigE", "tensorrt_llm::executor::ExecutorConfig::mCacheTransceiverConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig12mDebugConfigE", "tensorrt_llm::executor::ExecutorConfig::mDebugConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mEnableChunkedContextE", "tensorrt_llm::executor::ExecutorConfig::mEnableChunkedContext"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mEnableTrtOverlapE", "tensorrt_llm::executor::ExecutorConfig::mEnableTrtOverlap"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig30mExtendedRuntimePerfKnobConfigE", "tensorrt_llm::executor::ExecutorConfig::mExtendedRuntimePerfKnobConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mGatherGenerationLogitsE", "tensorrt_llm::executor::ExecutorConfig::mGatherGenerationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mGpuWeightsPercentE", "tensorrt_llm::executor::ExecutorConfig::mGpuWeightsPercent"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig21mGuidedDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mGuidedDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mIterStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::mIterStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14mKvCacheConfigE", "tensorrt_llm::executor::ExecutorConfig::mKvCacheConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mLogitsPostProcessorConfigE", "tensorrt_llm::executor::ExecutorConfig::mLogitsPostProcessorConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBatchSizeE", "tensorrt_llm::executor::ExecutorConfig::mMaxBatchSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxBeamWidthE", "tensorrt_llm::executor::ExecutorConfig::mMaxBeamWidth"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxNumTokensE", "tensorrt_llm::executor::ExecutorConfig::mMaxNumTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig13mMaxQueueSizeE", "tensorrt_llm::executor::ExecutorConfig::mMaxQueueSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23mMaxSeqIdleMicrosecondsE", "tensorrt_llm::executor::ExecutorConfig::mMaxSeqIdleMicroseconds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18mNormalizeLogProbsE", "tensorrt_llm::executor::ExecutorConfig::mNormalizeLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15mParallelConfigE", "tensorrt_llm::executor::ExecutorConfig::mParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mPeftCacheConfigE", "tensorrt_llm::executor::ExecutorConfig::mPeftCacheConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22mPromptTableOffloadingE", "tensorrt_llm::executor::ExecutorConfig::mPromptTableOffloading"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17mRecvPollPeriodMsE", "tensorrt_llm::executor::ExecutorConfig::mRecvPollPeriodMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mRequestStatsMaxIterationsE", "tensorrt_llm::executor::ExecutorConfig::mRequestStatsMaxIterations"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16mSchedulerConfigE", "tensorrt_llm::executor::ExecutorConfig::mSchedulerConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig26mSpeculativeDecodingConfigE", "tensorrt_llm::executor::ExecutorConfig::mSpeculativeDecodingConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20mUseGpuDirectStorageE", "tensorrt_llm::executor::ExecutorConfig::mUseGpuDirectStorage"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", "tensorrt_llm::executor::ExecutorConfig::setAdditionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setAdditionalModelOutputsERKNSt6vectorI21AdditionalModelOutputEE", "tensorrt_llm::executor::ExecutorConfig::setAdditionalModelOutputs::additionalModelOutputs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", "tensorrt_llm::executor::ExecutorConfig::setBatchingType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setBatchingTypeE12BatchingType", "tensorrt_llm::executor::ExecutorConfig::setBatchingType::batchingType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", "tensorrt_llm::executor::ExecutorConfig::setCacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setCacheTransceiverConfigERK22CacheTransceiverConfig", "tensorrt_llm::executor::ExecutorConfig::setCacheTransceiverConfig::cacheTransceiverConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", "tensorrt_llm::executor::ExecutorConfig::setDebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig14setDebugConfigERK11DebugConfig", "tensorrt_llm::executor::ExecutorConfig::setDebugConfig::debugConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setDecodingConfigERK14DecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setDecodingConfig::decodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", "tensorrt_llm::executor::ExecutorConfig::setEnableChunkedContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setEnableChunkedContextEb", "tensorrt_llm::executor::ExecutorConfig::setEnableChunkedContext::enableChunkedContext"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", "tensorrt_llm::executor::ExecutorConfig::setEnableTrtOverlap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setEnableTrtOverlapEb", "tensorrt_llm::executor::ExecutorConfig::setEnableTrtOverlap::enableTrtOverlap"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExecutorConfig::setExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig32setExtendedRuntimePerfKnobConfigERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExecutorConfig::setExtendedRuntimePerfKnobConfig::extendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", "tensorrt_llm::executor::ExecutorConfig::setGatherGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setGatherGenerationLogitsEb", "tensorrt_llm::executor::ExecutorConfig::setGatherGenerationLogits::gatherGenerationLogits"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", "tensorrt_llm::executor::ExecutorConfig::setGpuWeightsPercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setGpuWeightsPercentERKf", "tensorrt_llm::executor::ExecutorConfig::setGpuWeightsPercent::gpuWeightsPercent"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setGuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig23setGuidedDecodingConfigERK20GuidedDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setGuidedDecodingConfig::guidedDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setIterStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setIterStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setIterStatsMaxIterations::iterStatsMaxIterations"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setKvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setKvCacheConfigERK13KvCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setKvCacheConfig::kvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", "tensorrt_llm::executor::ExecutorConfig::setLogitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setLogitsPostProcessorConfigERK25LogitsPostProcessorConfig", "tensorrt_llm::executor::ExecutorConfig::setLogitsPostProcessorConfig::logitsPostProcessorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBatchSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBatchSize::maxBatchSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxBeamWidth::maxBeamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxNumTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxNumTokensE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setMaxNumTokens::maxNumTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ExecutorConfig::setMaxQueueSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig15setMaxQueueSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ExecutorConfig::setMaxQueueSize::maxQueueSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", "tensorrt_llm::executor::ExecutorConfig::setMaxSeqIdleMicroseconds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig25setMaxSeqIdleMicrosecondsE8uint64_t", "tensorrt_llm::executor::ExecutorConfig::setMaxSeqIdleMicroseconds::maxSeqIdleMicroseconds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", "tensorrt_llm::executor::ExecutorConfig::setNormalizeLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig20setNormalizeLogProbsEb", "tensorrt_llm::executor::ExecutorConfig::setNormalizeLogProbs::normalizeLogProbs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", "tensorrt_llm::executor::ExecutorConfig::setParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig17setParallelConfigERK14ParallelConfig", "tensorrt_llm::executor::ExecutorConfig::setParallelConfig::parallelConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setPeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setPeftCacheConfigERK15PeftCacheConfig", "tensorrt_llm::executor::ExecutorConfig::setPeftCacheConfig::peftCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", "tensorrt_llm::executor::ExecutorConfig::setPromptTableOffloading"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig24setPromptTableOffloadingEb", "tensorrt_llm::executor::ExecutorConfig::setPromptTableOffloading::promptTableOffloading"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRecvPollPeriodMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig19setRecvPollPeriodMsERK10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRecvPollPeriodMs::recvPollPeriodMs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRequestStatsMaxIterations"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig28setRequestStatsMaxIterationsE10SizeType32", "tensorrt_llm::executor::ExecutorConfig::setRequestStatsMaxIterations::requestStatsMaxIterations"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", "tensorrt_llm::executor::ExecutorConfig::setSchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig18setSchedulerConfigERK15SchedulerConfig", "tensorrt_llm::executor::ExecutorConfig::setSchedulerConfig::schedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setSpecDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig16setSpecDecConfigERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::ExecutorConfig::setSpecDecConfig::specDecConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", "tensorrt_llm::executor::ExecutorConfig::setUseGpuDirectStorage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ExecutorConfig22setUseGpuDirectStorageERKb", "tensorrt_llm::executor::ExecutorConfig::setUseGpuDirectStorage::useGpuDirectStorage"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::cudaGraphCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::cudaGraphMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::enableContextFMHAFP32Acc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig29ExtendedRuntimePerfKnobConfigEbbb10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::ExtendedRuntimePerfKnobConfig::multiBlockMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21getCudaGraphCacheSizeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getCudaGraphCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16getCudaGraphModeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getCudaGraphMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27getEnableContextFMHAFP32AccEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getEnableContextFMHAFP32Acc"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17getMultiBlockModeEv", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::getMultiBlockMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig19mCudaGraphCacheSizeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mCudaGraphCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig14mCudaGraphModeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mCudaGraphMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig25mEnableContextFMHAFP32AccE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mEnableContextFMHAFP32Acc"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig15mMultiBlockModeE", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::mMultiBlockMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfigeqERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig21setCudaGraphCacheSizeE10SizeType32", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphCacheSize::cacheSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig16setCudaGraphModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setCudaGraphMode::cudaGraphMode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setEnableContextFMHAFP32Acc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig27setEnableContextFMHAFP32AccEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setEnableContextFMHAFP32Acc::enableContextFMHAFP32Acc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setMultiBlockMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor29ExtendedRuntimePerfKnobConfig17setMultiBlockModeEb", "tensorrt_llm::executor::ExtendedRuntimePerfKnobConfig::setMultiBlockMode::multiBlockMode"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfigE", "tensorrt_llm::executor::ExternalDraftTokensConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::acceptanceThreshold"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::fastLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::logits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig25ExternalDraftTokensConfigE9VecTokensNSt8optionalI6TensorEERKNSt8optionalI9FloatTypeEERKNSt8optionalIbEE", "tensorrt_llm::executor::ExternalDraftTokensConfig::ExternalDraftTokensConfig::tokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig22getAcceptanceThresholdEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getAcceptanceThreshold"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig13getFastLogitsEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getFastLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getLogitsEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25ExternalDraftTokensConfig9getTokensEv", "tensorrt_llm::executor::ExternalDraftTokensConfig::getTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig20mAcceptanceThresholdE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mAcceptanceThreshold"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig11mFastLogitsE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mFastLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mLogitsE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25ExternalDraftTokensConfig7mTokensE", "tensorrt_llm::executor::ExternalDraftTokensConfig::mTokens"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12FinishReasonE", "tensorrt_llm::executor::FinishReason"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason10kCANCELLEDE", "tensorrt_llm::executor::FinishReason::kCANCELLED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason7kEND_IDE", "tensorrt_llm::executor::FinishReason::kEND_ID"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason7kLENGTHE", "tensorrt_llm::executor::FinishReason::kLENGTH"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason13kNOT_FINISHEDE", "tensorrt_llm::executor::FinishReason::kNOT_FINISHED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason11kSTOP_WORDSE", "tensorrt_llm::executor::FinishReason::kSTOP_WORDS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12FinishReason10kTIMED_OUTE", "tensorrt_llm::executor::FinishReason::kTIMED_OUT"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9FloatTypeE", "tensorrt_llm::executor::FloatType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfigE", "tensorrt_llm::executor::GuidedDecodingConfig"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackendE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingBackend"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig21GuidedDecodingBackend9kXGRAMMARE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingBackend::kXGRAMMAR"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::backend"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::encodedVocab"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::stopTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig20GuidedDecodingConfigE21GuidedDecodingBackendNSt8optionalINSt6vectorINSt6stringEEEEENSt8optionalINSt6stringEEENSt8optionalINSt6vectorI11TokenIdTypeEEEE", "tensorrt_llm::executor::GuidedDecodingConfig::GuidedDecodingConfig::tokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig10getBackendEv", "tensorrt_llm::executor::GuidedDecodingConfig::getBackend"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getEncodedVocabEv", "tensorrt_llm::executor::GuidedDecodingConfig::getEncodedVocab"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getStopTokenIdsEv", "tensorrt_llm::executor::GuidedDecodingConfig::getStopTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig15getTokenizerStrEv", "tensorrt_llm::executor::GuidedDecodingConfig::getTokenizerStr"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig8mBackendE", "tensorrt_llm::executor::GuidedDecodingConfig::mBackend"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mEncodedVocabE", "tensorrt_llm::executor::GuidedDecodingConfig::mEncodedVocab"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mStopTokenIdsE", "tensorrt_llm::executor::GuidedDecodingConfig::mStopTokenIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig13mTokenizerStrE", "tensorrt_llm::executor::GuidedDecodingConfig::mTokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", "tensorrt_llm::executor::GuidedDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfigeqERK20GuidedDecodingConfig", "tensorrt_llm::executor::GuidedDecodingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", "tensorrt_llm::executor::GuidedDecodingConfig::setBackend"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig10setBackendERK21GuidedDecodingBackend", "tensorrt_llm::executor::GuidedDecodingConfig::setBackend::backend"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingConfig::setEncodedVocab"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setEncodedVocabERKNSt6vectorINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingConfig::setEncodedVocab::encodedVocab"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", "tensorrt_llm::executor::GuidedDecodingConfig::setStopTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setStopTokenIdsERKNSt6vectorI11TokenIdTypeEE", "tensorrt_llm::executor::GuidedDecodingConfig::setStopTokenIds::stopTokenIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", "tensorrt_llm::executor::GuidedDecodingConfig::setTokenizerStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingConfig15setTokenizerStrERKNSt6stringE", "tensorrt_llm::executor::GuidedDecodingConfig::setTokenizerStr::tokenizerStr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingConfig8validateEv", "tensorrt_llm::executor::GuidedDecodingConfig::validate"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParamsE", "tensorrt_llm::executor::GuidedDecodingParams"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideTypeE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType13kEBNF_GRAMMARE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kEBNF_GRAMMAR"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType5kJSONE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kJSON"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType12kJSON_SCHEMAE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kJSON_SCHEMA"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType6kREGEXE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kREGEX"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams9GuideType15kSTRUCTURAL_TAGE", "tensorrt_llm::executor::GuidedDecodingParams::GuideType::kSTRUCTURAL_TAG"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams::guide"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams20GuidedDecodingParamsE9GuideTypeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::GuidedDecodingParams::GuidedDecodingParams::guideType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams8getGuideEv", "tensorrt_llm::executor::GuidedDecodingParams::getGuide"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParams12getGuideTypeEv", "tensorrt_llm::executor::GuidedDecodingParams::getGuideType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams6mGuideE", "tensorrt_llm::executor::GuidedDecodingParams::mGuide"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor20GuidedDecodingParams10mGuideTypeE", "tensorrt_llm::executor::GuidedDecodingParams::mGuideType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", "tensorrt_llm::executor::GuidedDecodingParams::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor20GuidedDecodingParamseqERK20GuidedDecodingParams", "tensorrt_llm::executor::GuidedDecodingParams::operator==::other"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6IdTypeE", "tensorrt_llm::executor::IdType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStatsE", "tensorrt_llm::executor::InflightBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats26avgNumDecodedTokensPerIterE", "tensorrt_llm::executor::InflightBatchingStats::avgNumDecodedTokensPerIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12microBatchIdE", "tensorrt_llm::executor::InflightBatchingStats::microBatchId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats18numContextRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numContextRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats12numCtxTokensE", "tensorrt_llm::executor::InflightBatchingStats::numCtxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats14numGenRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numGenRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats17numPausedRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numPausedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor21InflightBatchingStats20numScheduledRequestsE", "tensorrt_llm::executor::InflightBatchingStats::numScheduledRequests"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14IterationStatsE", "tensorrt_llm::executor::IterationStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats11cpuMemUsageE", "tensorrt_llm::executor::IterationStats::cpuMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17crossKvCacheStatsE", "tensorrt_llm::executor::IterationStats::crossKvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats11gpuMemUsageE", "tensorrt_llm::executor::IterationStats::gpuMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats21inflightBatchingStatsE", "tensorrt_llm::executor::IterationStats::inflightBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats4iterE", "tensorrt_llm::executor::IterationStats::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats13iterLatencyMSE", "tensorrt_llm::executor::IterationStats::iterLatencyMS"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats12kvCacheStatsE", "tensorrt_llm::executor::IterationStats::kvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxBatchSizeRuntimeE", "tensorrt_llm::executor::IterationStats::maxBatchSizeRuntime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxBatchSizeStaticE", "tensorrt_llm::executor::IterationStats::maxBatchSizeStatic"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxBatchSizeTunerRecommendedE", "tensorrt_llm::executor::IterationStats::maxBatchSizeTunerRecommended"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20maxNumActiveRequestsE", "tensorrt_llm::executor::IterationStats::maxNumActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19maxNumTokensRuntimeE", "tensorrt_llm::executor::IterationStats::maxNumTokensRuntime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats18maxNumTokensStaticE", "tensorrt_llm::executor::IterationStats::maxNumTokensStatic"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats28maxNumTokensTunerRecommendedE", "tensorrt_llm::executor::IterationStats::maxNumTokensTunerRecommended"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats31newActiveRequestsQueueLatencyMSE", "tensorrt_llm::executor::IterationStats::newActiveRequestsQueueLatencyMS"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17numActiveRequestsE", "tensorrt_llm::executor::IterationStats::numActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20numCompletedRequestsE", "tensorrt_llm::executor::IterationStats::numCompletedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats20numNewActiveRequestsE", "tensorrt_llm::executor::IterationStats::numNewActiveRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats17numQueuedRequestsE", "tensorrt_llm::executor::IterationStats::numQueuedRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats14pinnedMemUsageE", "tensorrt_llm::executor::IterationStats::pinnedMemUsage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats12specDecStatsE", "tensorrt_llm::executor::IterationStats::specDecStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats19staticBatchingStatsE", "tensorrt_llm::executor::IterationStats::staticBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14IterationStats9timestampE", "tensorrt_llm::executor::IterationStats::timestamp"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor13IterationTypeE", "tensorrt_llm::executor::IterationType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerializationE", "tensorrt_llm::executor::JsonSerialization"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", "tensorrt_llm::executor::JsonSerialization::toJsonStr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK14IterationStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr::iterationStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK12RequestStats", "tensorrt_llm::executor::JsonSerialization::toJsonStr::requestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor17JsonSerialization9toJsonStrERK24RequestStatsPerIteration", "tensorrt_llm::executor::JsonSerialization::toJsonStr::requestStatsPerIter"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedDataE", "tensorrt_llm::executor::KVCacheCreatedData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheCreatedData22numBlocksPerCacheLevelE", "tensorrt_llm::executor::KVCacheCreatedData::numBlocksPerCacheLevel"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEventE", "tensorrt_llm::executor::KVCacheEvent"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent12KVCacheEventE6IdType16KVCacheEventData", "tensorrt_llm::executor::KVCacheEvent::KVCacheEvent::eventId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent4dataE", "tensorrt_llm::executor::KVCacheEvent::data"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KVCacheEvent7eventIdE", "tensorrt_llm::executor::KVCacheEvent::eventId"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDataE", "tensorrt_llm::executor::KVCacheEventData"], [0, 2, 1, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", "tensorrt_llm::executor::KVCacheEventDiff"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor16KVCacheEventDiffE", "tensorrt_llm::executor::KVCacheEventDiff::T"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8newValueE", "tensorrt_llm::executor::KVCacheEventDiff::newValue"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor16KVCacheEventDiff8oldValueE", "tensorrt_llm::executor::KVCacheEventDiff::oldValue"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManagerE", "tensorrt_llm::executor::KVCacheEventManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", "tensorrt_llm::executor::KVCacheEventManager::KVCacheEventManager"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager19KVCacheEventManagerENSt10shared_ptrIN12tensorrt_llm13batch_manager16kv_cache_manager18BaseKVCacheManagerEEE", "tensorrt_llm::executor::KVCacheEventManager::KVCacheEventManager::kvCacheManager"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KVCacheEventManager::getLatestEvents"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager15getLatestEventsENSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KVCacheEventManager::getLatestEvents::timeout"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19KVCacheEventManager14kvCacheManagerE", "tensorrt_llm::executor::KVCacheEventManager::kvCacheManager"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedDataE", "tensorrt_llm::executor::KVCacheRemovedData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheRemovedData11blockHashesE", "tensorrt_llm::executor::KVCacheRemovedData::blockHashes"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockDataE", "tensorrt_llm::executor::KVCacheStoredBlockData"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::blockHash"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::cacheLevel"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::loraId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData22KVCacheStoredBlockDataE6IdTypeN12tensorrt_llm7runtime15VecUniqueTokensENSt8optionalIN12tensorrt_llm7runtime14LoraTaskIdTypeEEE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheStoredBlockData::KVCacheStoredBlockData::tokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData9blockHashE", "tensorrt_llm::executor::KVCacheStoredBlockData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData10cacheLevelE", "tensorrt_llm::executor::KVCacheStoredBlockData::cacheLevel"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6loraIdE", "tensorrt_llm::executor::KVCacheStoredBlockData::loraId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData8priorityE", "tensorrt_llm::executor::KVCacheStoredBlockData::priority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KVCacheStoredBlockData6tokensE", "tensorrt_llm::executor::KVCacheStoredBlockData::tokens"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredDataE", "tensorrt_llm::executor::KVCacheStoredData"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData6blocksE", "tensorrt_llm::executor::KVCacheStoredData::blocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17KVCacheStoredData10parentHashE", "tensorrt_llm::executor::KVCacheStoredData::parentHash"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedDataE", "tensorrt_llm::executor::KVCacheUpdatedData"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", "tensorrt_llm::executor::KVCacheUpdatedData::KVCacheUpdatedData"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData18KVCacheUpdatedDataE6IdType", "tensorrt_llm::executor::KVCacheUpdatedData::KVCacheUpdatedData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData9blockHashE", "tensorrt_llm::executor::KVCacheUpdatedData::blockHash"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData10cacheLevelE", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevel"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated::newValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData17cacheLevelUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::cacheLevelUpdated::oldValue"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData8priorityE", "tensorrt_llm::executor::KVCacheUpdatedData::priority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated::newValue"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18KVCacheUpdatedData15priorityUpdatedE10SizeType3210SizeType32", "tensorrt_llm::executor::KVCacheUpdatedData::priorityUpdated::oldValue"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfigE", "tensorrt_llm::executor::KvCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::copyOnPartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::crossKvCacheFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::enableBlockReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::enablePartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::eventBufferMaxSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::freeGpuMemoryFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::hostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::maxAttentionWindowVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::maxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::onboardBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::runtimeDefaults"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::secondaryOffloadMinPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig13KvCacheConfigEbRKNSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI6size_tEEbRKNSt8optionalI9FloatTypeEENSt8optionalI17RetentionPriorityEE6size_tRKNSt8optionalIN12tensorrt_llm7runtime15RuntimeDefaultsEEEbb", "tensorrt_llm::executor::KvCacheConfig::KvCacheConfig::sinkTokenLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::executor::KvCacheConfig::fillEmptyFieldsFromRuntimeDefaults"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig34fillEmptyFieldsFromRuntimeDefaultsEN12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::executor::KvCacheConfig::fillEmptyFieldsFromRuntimeDefaults::runtimeDefaults"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getCopyOnPartialReuseEv", "tensorrt_llm::executor::KvCacheConfig::getCopyOnPartialReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig23getCrossKvCacheFractionEv", "tensorrt_llm::executor::KvCacheConfig::getCrossKvCacheFraction"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig19getEnableBlockReuseEv", "tensorrt_llm::executor::KvCacheConfig::getEnableBlockReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEnablePartialReuseEv", "tensorrt_llm::executor::KvCacheConfig::getEnablePartialReuse"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig21getEventBufferMaxSizeEv", "tensorrt_llm::executor::KvCacheConfig::getEventBufferMaxSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getFreeGpuMemoryFractionEv", "tensorrt_llm::executor::KvCacheConfig::getFreeGpuMemoryFraction"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getHostCacheSizeEv", "tensorrt_llm::executor::KvCacheConfig::getHostCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig24getMaxAttentionWindowVecEv", "tensorrt_llm::executor::KvCacheConfig::getMaxAttentionWindowVec"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig12getMaxTokensEv", "tensorrt_llm::executor::KvCacheConfig::getMaxTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig16getOnboardBlocksEv", "tensorrt_llm::executor::KvCacheConfig::getOnboardBlocks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig30getSecondaryOffloadMinPriorityEv", "tensorrt_llm::executor::KvCacheConfig::getSecondaryOffloadMinPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor13KvCacheConfig18getSinkTokenLengthEv", "tensorrt_llm::executor::KvCacheConfig::getSinkTokenLength"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mCopyOnPartialReuseE", "tensorrt_llm::executor::KvCacheConfig::mCopyOnPartialReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21mCrossKvCacheFractionE", "tensorrt_llm::executor::KvCacheConfig::mCrossKvCacheFraction"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig17mEnableBlockReuseE", "tensorrt_llm::executor::KvCacheConfig::mEnableBlockReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEnablePartialReuseE", "tensorrt_llm::executor::KvCacheConfig::mEnablePartialReuse"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19mEventBufferMaxSizeE", "tensorrt_llm::executor::KvCacheConfig::mEventBufferMaxSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mFreeGpuMemoryFractionE", "tensorrt_llm::executor::KvCacheConfig::mFreeGpuMemoryFraction"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mHostCacheSizeE", "tensorrt_llm::executor::KvCacheConfig::mHostCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig22mMaxAttentionWindowVecE", "tensorrt_llm::executor::KvCacheConfig::mMaxAttentionWindowVec"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig10mMaxTokensE", "tensorrt_llm::executor::KvCacheConfig::mMaxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig14mOnboardBlocksE", "tensorrt_llm::executor::KvCacheConfig::mOnboardBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig28mSecondaryOffloadMinPriorityE", "tensorrt_llm::executor::KvCacheConfig::mSecondaryOffloadMinPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16mSinkTokenLengthE", "tensorrt_llm::executor::KvCacheConfig::mSinkTokenLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setCopyOnPartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setCopyOnPartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setCopyOnPartialReuse::copyOnPartialReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setCrossKvCacheFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig23setCrossKvCacheFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setCrossKvCacheFraction::crossKvCacheFraction"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnableBlockReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig19setEnableBlockReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnableBlockReuse::enableBlockReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnablePartialReuse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEnablePartialReuseEb", "tensorrt_llm::executor::KvCacheConfig::setEnablePartialReuse::enablePartialReuse"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setEventBufferMaxSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig21setEventBufferMaxSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setEventBufferMaxSize::eventBufferMaxSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setFreeGpuMemoryFraction"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setFreeGpuMemoryFractionE9FloatType", "tensorrt_llm::executor::KvCacheConfig::setFreeGpuMemoryFraction::freeGpuMemoryFraction"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setHostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setHostCacheSizeE6size_t", "tensorrt_llm::executor::KvCacheConfig::setHostCacheSize::hostCacheSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", "tensorrt_llm::executor::KvCacheConfig::setMaxAttentionWindowVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig24setMaxAttentionWindowVecENSt6vectorI10SizeType32EE", "tensorrt_llm::executor::KvCacheConfig::setMaxAttentionWindowVec::maxAttentionWindowVec"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setMaxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig12setMaxTokensE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setMaxTokens::maxTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", "tensorrt_llm::executor::KvCacheConfig::setOnboardBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig16setOnboardBlocksEb", "tensorrt_llm::executor::KvCacheConfig::setOnboardBlocks::onboardBlocks"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", "tensorrt_llm::executor::KvCacheConfig::setSecondaryOffloadMinPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig30setSecondaryOffloadMinPriorityENSt8optionalI17RetentionPriorityEE", "tensorrt_llm::executor::KvCacheConfig::setSecondaryOffloadMinPriority::secondaryOffloadMinPriority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setSinkTokenLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13KvCacheConfig18setSinkTokenLengthE10SizeType32", "tensorrt_llm::executor::KvCacheConfig::setSinkTokenLength::sinkTokenLength"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfigE", "tensorrt_llm::executor::KvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigEv", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::decodeDurationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::decodeRetentionPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::directory"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::tokenRangeRetentionPriorities"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig22KvCacheRetentionConfigERKNSt6vectorI25TokenRangeRetentionConfigEE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE19KvCacheTransferModeNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::KvCacheRetentionConfig::transferMode"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::durationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::tokenEnd"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig25TokenRangeRetentionConfigE10SizeType32NSt8optionalI10SizeType32EE17RetentionPriorityNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::TokenRangeRetentionConfig::tokenStart"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10durationMsE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::durationMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfigeqERK25TokenRangeRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::operator==::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8priorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::priority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig8tokenEndE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::tokenEnd"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25TokenRangeRetentionConfig10tokenStartE", "tensorrt_llm::executor::KvCacheRetentionConfig::TokenRangeRetentionConfig::tokenStart"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig19getDecodeDurationMsEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeDurationMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig26getDecodeRetentionPriorityEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getDecodeRetentionPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig12getDirectoryEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getDirectory"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration::blockSize"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig36getPerBlockRetentionPriorityDurationE10SizeType3210SizeType32", "tensorrt_llm::executor::KvCacheRetentionConfig::getPerBlockRetentionPriorityDuration::seqLen"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig29getTokenRangeRetentionConfigsEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getTokenRangeRetentionConfigs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfig15getTransferModeEv", "tensorrt_llm::executor::KvCacheRetentionConfig::getTransferMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig25kDefaultRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kDefaultRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMaxRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kMaxRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig21kMinRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::kMinRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig17mDecodeDurationMsE", "tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeDurationMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig24mDecodeRetentionPriorityE", "tensorrt_llm::executor::KvCacheRetentionConfig::mDecodeRetentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig10mDirectoryE", "tensorrt_llm::executor::KvCacheRetentionConfig::mDirectory"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig27mTokenRangeRetentionConfigsE", "tensorrt_llm::executor::KvCacheRetentionConfig::mTokenRangeRetentionConfigs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor22KvCacheRetentionConfig13mTransferModeE", "tensorrt_llm::executor::KvCacheRetentionConfig::mTransferMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor22KvCacheRetentionConfigeqERK22KvCacheRetentionConfig", "tensorrt_llm::executor::KvCacheRetentionConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStatsE", "tensorrt_llm::executor::KvCacheStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14allocNewBlocksE", "tensorrt_llm::executor::KvCacheStats::allocNewBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats16allocTotalBlocksE", "tensorrt_llm::executor::KvCacheStats::allocTotalBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12cacheHitRateE", "tensorrt_llm::executor::KvCacheStats::cacheHitRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13freeNumBlocksE", "tensorrt_llm::executor::KvCacheStats::freeNumBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12maxNumBlocksE", "tensorrt_llm::executor::KvCacheStats::maxNumBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12missedBlocksE", "tensorrt_llm::executor::KvCacheStats::missedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats12reusedBlocksE", "tensorrt_llm::executor::KvCacheStats::reusedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats14tokensPerBlockE", "tensorrt_llm::executor::KvCacheStats::tokensPerBlock"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12KvCacheStats13usedNumBlocksE", "tensorrt_llm::executor::KvCacheStats::usedNumBlocks"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferModeE", "tensorrt_llm::executor::KvCacheTransferMode"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode4DRAME", "tensorrt_llm::executor::KvCacheTransferMode::DRAM"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode3GDSE", "tensorrt_llm::executor::KvCacheTransferMode::GDS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor19KvCacheTransferMode20POSIX_DEBUG_FALLBACKE", "tensorrt_llm::executor::KvCacheTransferMode::POSIX_DEBUG_FALLBACK"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor19LogitsPostProcessorE", "tensorrt_llm::executor::LogitsPostProcessor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor26LogitsPostProcessorBatchedE", "tensorrt_llm::executor::LogitsPostProcessorBatched"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfigE", "tensorrt_llm::executor::LogitsPostProcessorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::processorBatched"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::processorMap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig25LogitsPostProcessorConfigENSt8optionalI22LogitsPostProcessorMapEENSt8optionalI26LogitsPostProcessorBatchedEEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::LogitsPostProcessorConfig::replicate"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig19getProcessorBatchedEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getProcessorBatched"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig15getProcessorMapEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getProcessorMap"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25LogitsPostProcessorConfig12getReplicateEv", "tensorrt_llm::executor::LogitsPostProcessorConfig::getReplicate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig17mProcessorBatchedE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mProcessorBatched"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig13mProcessorMapE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mProcessorMap"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig10mReplicateE", "tensorrt_llm::executor::LogitsPostProcessorConfig::mReplicate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorBatched"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig19setProcessorBatchedERK26LogitsPostProcessorBatched", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorBatched::processorBatched"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorMap"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig15setProcessorMapERK22LogitsPostProcessorMap", "tensorrt_llm::executor::LogitsPostProcessorConfig::setProcessorMap::processorMap"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::setReplicate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25LogitsPostProcessorConfig12setReplicateEb", "tensorrt_llm::executor::LogitsPostProcessorConfig::setReplicate::replicate"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor22LogitsPostProcessorMapE", "tensorrt_llm::executor::LogitsPostProcessorMap"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfigE", "tensorrt_llm::executor::LookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigEv", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig23LookaheadDecodingConfigE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::LookaheadDecodingConfig::windowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig28calculateSpeculativeResourceEv", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResource"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig33calculateSpeculativeResourceTupleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::calculateSpeculativeResourceTuple::windowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig3getEv", "tensorrt_llm::executor::LookaheadDecodingConfig::get"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig12getNgramSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig22getVerificationSetSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getVerificationSetSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig13getWindowSizeEv", "tensorrt_llm::executor::LookaheadDecodingConfig::getWindowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::isLE"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfig4isLEERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::isLE::that"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::ngramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::verificationSetSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig7isLegalE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::executor::LookaheadDecodingConfig::isLegal::windowSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig30kDefaultLookaheadDecodingNgramE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingNgram"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig40kDefaultLookaheadDecodingVerificationSetE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingVerificationSet"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig31kDefaultLookaheadDecodingWindowE", "tensorrt_llm::executor::LookaheadDecodingConfig::kDefaultLookaheadDecodingWindow"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig10mNgramSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig20mVerificationSetSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mVerificationSetSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor23LookaheadDecodingConfig11mWindowSizeE", "tensorrt_llm::executor::LookaheadDecodingConfig::mWindowSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor23LookaheadDecodingConfigeqERK23LookaheadDecodingConfig", "tensorrt_llm::executor::LookaheadDecodingConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfigE", "tensorrt_llm::executor::LoraConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::taskId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig10LoraConfigE6IdTypeNSt8optionalI6TensorEENSt8optionalI6TensorEE", "tensorrt_llm::executor::LoraConfig::LoraConfig::weights"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getConfigEv", "tensorrt_llm::executor::LoraConfig::getConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig9getTaskIdEv", "tensorrt_llm::executor::LoraConfig::getTaskId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor10LoraConfig10getWeightsEv", "tensorrt_llm::executor::LoraConfig::getWeights"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mConfigE", "tensorrt_llm::executor::LoraConfig::mConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig7mTaskIdE", "tensorrt_llm::executor::LoraConfig::mTaskId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10LoraConfig8mWeightsE", "tensorrt_llm::executor::LoraConfig::mWeights"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor13MedusaChoicesE", "tensorrt_llm::executor::MedusaChoices"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor10MemoryTypeE", "tensorrt_llm::executor::MemoryType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kCPUE", "tensorrt_llm::executor::MemoryType::kCPU"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType11kCPU_PINNEDE", "tensorrt_llm::executor::MemoryType::kCPU_PINNED"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType15kCPU_PINNEDPOOLE", "tensorrt_llm::executor::MemoryType::kCPU_PINNEDPOOL"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kGPUE", "tensorrt_llm::executor::MemoryType::kGPU"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType8kUNKNOWNE", "tensorrt_llm::executor::MemoryType::kUNKNOWN"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor10MemoryType4kUVME", "tensorrt_llm::executor::MemoryType::kUVM"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16MillisecondsTypeE", "tensorrt_llm::executor::MillisecondsType"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor9ModelTypeE", "tensorrt_llm::executor::ModelType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType13kDECODER_ONLYE", "tensorrt_llm::executor::ModelType::kDECODER_ONLY"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType16kENCODER_DECODERE", "tensorrt_llm::executor::ModelType::kENCODER_DECODER"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor9ModelType13kENCODER_ONLYE", "tensorrt_llm::executor::ModelType::kENCODER_ONLY"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfigE", "tensorrt_llm::executor::MropeConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig::mropePositionDeltas"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig11MropeConfigE6Tensor10SizeType32", "tensorrt_llm::executor::MropeConfig::MropeConfig::mropeRoratySinCos"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11MropeConfig22getMRopePositionDeltasEv", "tensorrt_llm::executor::MropeConfig::getMRopePositionDeltas"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor11MropeConfig20getMRopeRotaryCosSinEv", "tensorrt_llm::executor::MropeConfig::getMRopeRotaryCosSin"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig20mMRopePositionDeltasE", "tensorrt_llm::executor::MropeConfig::mMRopePositionDeltas"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor11MropeConfig18mMRopeRotaryCosSinE", "tensorrt_llm::executor::MropeConfig::mMRopeRotaryCosSin"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfigE", "tensorrt_llm::executor::OrchestratorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::isOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::orchLeaderComm"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::spawnProcesses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig18OrchestratorConfigEbNSt6stringENSt10shared_ptrIN3mpi7MpiCommEEEb", "tensorrt_llm::executor::OrchestratorConfig::OrchestratorConfig::workerExecutablePath"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getIsOrchestratorEv", "tensorrt_llm::executor::OrchestratorConfig::getIsOrchestrator"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getOrchLeaderCommEv", "tensorrt_llm::executor::OrchestratorConfig::getOrchLeaderComm"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig17getSpawnProcessesEv", "tensorrt_llm::executor::OrchestratorConfig::getSpawnProcesses"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18OrchestratorConfig23getWorkerExecutablePathEv", "tensorrt_llm::executor::OrchestratorConfig::getWorkerExecutablePath"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mIsOrchestratorE", "tensorrt_llm::executor::OrchestratorConfig::mIsOrchestrator"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mOrchLeaderCommE", "tensorrt_llm::executor::OrchestratorConfig::mOrchLeaderComm"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig15mSpawnProcessesE", "tensorrt_llm::executor::OrchestratorConfig::mSpawnProcesses"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig21mWorkerExecutablePathE", "tensorrt_llm::executor::OrchestratorConfig::mWorkerExecutablePath"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", "tensorrt_llm::executor::OrchestratorConfig::setIsOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setIsOrchestratorEb", "tensorrt_llm::executor::OrchestratorConfig::setIsOrchestrator::isOrchestrator"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", "tensorrt_llm::executor::OrchestratorConfig::setOrchLeaderComm"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setOrchLeaderCommERKNSt10shared_ptrIN3mpi7MpiCommEEE", "tensorrt_llm::executor::OrchestratorConfig::setOrchLeaderComm::orchLeaderComm"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", "tensorrt_llm::executor::OrchestratorConfig::setSpawnProcesses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig17setSpawnProcessesEb", "tensorrt_llm::executor::OrchestratorConfig::setSpawnProcesses::spawnProcesses"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", "tensorrt_llm::executor::OrchestratorConfig::setWorkerExecutablePath"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18OrchestratorConfig23setWorkerExecutablePathERKNSt6stringE", "tensorrt_llm::executor::OrchestratorConfig::setWorkerExecutablePath::workerExecutablePath"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfigE", "tensorrt_llm::executor::OutputConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::additionalModelOutputs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::excludeInputFromOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnContextLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnEncoderOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnGenerationLogits"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnLogProbs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig12OutputConfigEbbbbbbNSt8optionalINSt6vectorI21AdditionalModelOutputEEEE", "tensorrt_llm::executor::OutputConfig::OutputConfig::returnPerfMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22additionalModelOutputsE", "tensorrt_llm::executor::OutputConfig::additionalModelOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22excludeInputFromOutputE", "tensorrt_llm::executor::OutputConfig::excludeInputFromOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnContextLogitsE", "tensorrt_llm::executor::OutputConfig::returnContextLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig19returnEncoderOutputE", "tensorrt_llm::executor::OutputConfig::returnEncoderOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig22returnGenerationLogitsE", "tensorrt_llm::executor::OutputConfig::returnGenerationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig14returnLogProbsE", "tensorrt_llm::executor::OutputConfig::returnLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12OutputConfig17returnPerfMetricsE", "tensorrt_llm::executor::OutputConfig::returnPerfMetrics"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfigE", "tensorrt_llm::executor::ParallelConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::commMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::commType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::deviceIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::numNodes"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig14ParallelConfigE17CommunicationType17CommunicationModeNSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt6vectorI10SizeType32EEEERKNSt8optionalI18OrchestratorConfigEENSt8optionalI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::ParallelConfig::participantIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationModeEv", "tensorrt_llm::executor::ParallelConfig::getCommunicationMode"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig20getCommunicationTypeEv", "tensorrt_llm::executor::ParallelConfig::getCommunicationType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig12getDeviceIdsEv", "tensorrt_llm::executor::ParallelConfig::getDeviceIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig11getNumNodesEv", "tensorrt_llm::executor::ParallelConfig::getNumNodes"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig21getOrchestratorConfigEv", "tensorrt_llm::executor::ParallelConfig::getOrchestratorConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14ParallelConfig17getParticipantIdsEv", "tensorrt_llm::executor::ParallelConfig::getParticipantIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommModeE", "tensorrt_llm::executor::ParallelConfig::mCommMode"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mCommTypeE", "tensorrt_llm::executor::ParallelConfig::mCommType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig10mDeviceIdsE", "tensorrt_llm::executor::ParallelConfig::mDeviceIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig9mNumNodesE", "tensorrt_llm::executor::ParallelConfig::mNumNodes"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig19mOrchestratorConfigE", "tensorrt_llm::executor::ParallelConfig::mOrchestratorConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig15mParticipantIdsE", "tensorrt_llm::executor::ParallelConfig::mParticipantIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", "tensorrt_llm::executor::ParallelConfig::setCommunicationMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationModeE17CommunicationMode", "tensorrt_llm::executor::ParallelConfig::setCommunicationMode::mode"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", "tensorrt_llm::executor::ParallelConfig::setCommunicationType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig20setCommunicationTypeE17CommunicationType", "tensorrt_llm::executor::ParallelConfig::setCommunicationType::type"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setDeviceIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig12setDeviceIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setDeviceIds::deviceIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", "tensorrt_llm::executor::ParallelConfig::setNumNodes"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig11setNumNodesE10SizeType32", "tensorrt_llm::executor::ParallelConfig::setNumNodes::numNodes"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", "tensorrt_llm::executor::ParallelConfig::setOrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig21setOrchestratorConfigERK18OrchestratorConfig", "tensorrt_llm::executor::ParallelConfig::setOrchestratorConfig::orchestratorConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setParticipantIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14ParallelConfig17setParticipantIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::ParallelConfig::setParticipantIds::participantIds"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfigE", "tensorrt_llm::executor::PeftCacheConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::deviceCachePercent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::hostCacheSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::loraPrefetchDir"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxAdapterSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxPagesPerBlockDevice"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::maxPagesPerBlockHost"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numCopyStreams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numDeviceModuleLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numEnsureWorkers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numHostModuleLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::numPutWorkers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15PeftCacheConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalIfEERKNSt8optionalI6size_tEERKNSt8optionalINSt6stringEEE", "tensorrt_llm::executor::PeftCacheConfig::PeftCacheConfig::optimalAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getDeviceCachePercentEv", "tensorrt_llm::executor::PeftCacheConfig::getDeviceCachePercent"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getHostCacheSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getHostCacheSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig18getLoraPrefetchDirEv", "tensorrt_llm::executor::PeftCacheConfig::getLoraPrefetchDir"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getMaxAdapterSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig25getMaxPagesPerBlockDeviceEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxPagesPerBlockDevice"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getMaxPagesPerBlockHostEv", "tensorrt_llm::executor::PeftCacheConfig::getMaxPagesPerBlockHost"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig17getNumCopyStreamsEv", "tensorrt_llm::executor::PeftCacheConfig::getNumCopyStreams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig23getNumDeviceModuleLayerEv", "tensorrt_llm::executor::PeftCacheConfig::getNumDeviceModuleLayer"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig19getNumEnsureWorkersEv", "tensorrt_llm::executor::PeftCacheConfig::getNumEnsureWorkers"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getNumHostModuleLayerEv", "tensorrt_llm::executor::PeftCacheConfig::getNumHostModuleLayer"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig16getNumPutWorkersEv", "tensorrt_llm::executor::PeftCacheConfig::getNumPutWorkers"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfig21getOptimalAdapterSizeEv", "tensorrt_llm::executor::PeftCacheConfig::getOptimalAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig22kDefaultMaxAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig30kDefaultMaxPagesPerBlockDeviceE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxPagesPerBlockDevice"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig28kDefaultMaxPagesPerBlockHostE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultMaxPagesPerBlockHost"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig26kDefaultOptimalAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::kDefaultOptimalAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mDeviceCachePercentE", "tensorrt_llm::executor::PeftCacheConfig::mDeviceCachePercent"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mHostCacheSizeE", "tensorrt_llm::executor::PeftCacheConfig::mHostCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig16mLoraPrefetchDirE", "tensorrt_llm::executor::PeftCacheConfig::mLoraPrefetchDir"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mMaxAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::mMaxAdapterSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig23mMaxPagesPerBlockDeviceE", "tensorrt_llm::executor::PeftCacheConfig::mMaxPagesPerBlockDevice"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mMaxPagesPerBlockHostE", "tensorrt_llm::executor::PeftCacheConfig::mMaxPagesPerBlockHost"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig15mNumCopyStreamsE", "tensorrt_llm::executor::PeftCacheConfig::mNumCopyStreams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig21mNumDeviceModuleLayerE", "tensorrt_llm::executor::PeftCacheConfig::mNumDeviceModuleLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig17mNumEnsureWorkersE", "tensorrt_llm::executor::PeftCacheConfig::mNumEnsureWorkers"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mNumHostModuleLayerE", "tensorrt_llm::executor::PeftCacheConfig::mNumHostModuleLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig14mNumPutWorkersE", "tensorrt_llm::executor::PeftCacheConfig::mNumPutWorkers"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15PeftCacheConfig19mOptimalAdapterSizeE", "tensorrt_llm::executor::PeftCacheConfig::mOptimalAdapterSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", "tensorrt_llm::executor::PeftCacheConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor15PeftCacheConfigeqERK15PeftCacheConfig", "tensorrt_llm::executor::PeftCacheConfig::operator==::other"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor12PriorityTypeE", "tensorrt_llm::executor::PriorityType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfigE", "tensorrt_llm::executor::PromptTuningConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig::embeddingTable"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig18PromptTuningConfigE6TensorNSt8optionalI16VecTokenExtraIdsEE", "tensorrt_llm::executor::PromptTuningConfig::PromptTuningConfig::inputTokenExtraIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig17getEmbeddingTableEv", "tensorrt_llm::executor::PromptTuningConfig::getEmbeddingTable"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor18PromptTuningConfig21getInputTokenExtraIdsEv", "tensorrt_llm::executor::PromptTuningConfig::getInputTokenExtraIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig15mEmbeddingTableE", "tensorrt_llm::executor::PromptTuningConfig::mEmbeddingTable"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18PromptTuningConfig19mInputTokenExtraIdsE", "tensorrt_llm::executor::PromptTuningConfig::mInputTokenExtraIds"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor14RandomSeedTypeE", "tensorrt_llm::executor::RandomSeedType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor7RequestE", "tensorrt_llm::executor::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", "tensorrt_llm::executor::Request::Request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", "tensorrt_llm::executor::Request::Request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::allottedTimeMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::badWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::crossAttentionMask"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::embeddingBias"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderInputFeatures"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderInputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::encoderOutputLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::endId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::externalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::inputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::languageAdapterUid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::logitsPostProcessor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::logitsPostProcessorName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::lookaheadConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::loraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::mRopeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::maxTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::multimodalEmbedding"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::numReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERK7Request", "tensorrt_llm::executor::Request::Request::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestERR7Request", "tensorrt_llm::executor::Request::Request::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::outputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::pTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::padId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::positionIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::priority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::returnAllGeneratedTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::samplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::skipCrossAttnBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::stopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::streaming"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request7RequestE9VecTokens10SizeType32bRK14SamplingConfigRK12OutputConfigRKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalINSt4listI9VecTokensEEEENSt8optionalI6TensorEENSt8optionalI25ExternalDraftTokensConfigEENSt8optionalI18PromptTuningConfigEENSt8optionalI6TensorEENSt8optionalI11MropeConfigEENSt8optionalI10LoraConfigEENSt8optionalI23LookaheadDecodingConfigEENSt8optionalI22KvCacheRetentionConfigEENSt8optionalINSt6stringEEENSt8optionalI19LogitsPostProcessorEENSt8optionalI9VecTokensEENSt8optionalI6IdTypeEEb12PriorityType11RequestTypeNSt8optionalI18ContextPhaseParamsEENSt8optionalI6TensorEENSt8optionalI10SizeType32EENSt8optionalI6TensorEE10SizeType32NSt8optionalI11EagleConfigEENSt8optionalI6TensorEENSt8optionalI20GuidedDecodingParamsEENSt8optionalI10SizeType32EENSt8optionalI16MillisecondsTypeEE", "tensorrt_llm::executor::Request::Request::type"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request24getAdditionalOutputNamesEv", "tensorrt_llm::executor::Request::getAdditionalOutputNames"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request17getAllottedTimeMsEv", "tensorrt_llm::executor::Request::getAllottedTimeMs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getBadWordsEv", "tensorrt_llm::executor::Request::getBadWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getClientIdEv", "tensorrt_llm::executor::Request::getClientId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getContextPhaseParamsEv", "tensorrt_llm::executor::Request::getContextPhaseParams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getCrossAttentionMaskEv", "tensorrt_llm::executor::Request::getCrossAttentionMask"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getEagleConfigEv", "tensorrt_llm::executor::Request::getEagleConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request16getEmbeddingBiasEv", "tensorrt_llm::executor::Request::getEmbeddingBias"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputFeaturesEv", "tensorrt_llm::executor::Request::getEncoderInputFeatures"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getEncoderInputTokenIdsEv", "tensorrt_llm::executor::Request::getEncoderInputTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getEncoderOutputLengthEv", "tensorrt_llm::executor::Request::getEncoderOutputLength"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request8getEndIdEv", "tensorrt_llm::executor::Request::getEndId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request28getExternalDraftTokensConfigEv", "tensorrt_llm::executor::Request::getExternalDraftTokensConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request23getGuidedDecodingParamsEv", "tensorrt_llm::executor::Request::getGuidedDecodingParams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request16getInputTokenIdsEv", "tensorrt_llm::executor::Request::getInputTokenIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request25getKvCacheRetentionConfigEv", "tensorrt_llm::executor::Request::getKvCacheRetentionConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getLanguageAdapterUidEv", "tensorrt_llm::executor::Request::getLanguageAdapterUid"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getLogitsPostProcessorEv", "tensorrt_llm::executor::Request::getLogitsPostProcessor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request26getLogitsPostProcessorNameEv", "tensorrt_llm::executor::Request::getLogitsPostProcessorName"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request18getLookaheadConfigEv", "tensorrt_llm::executor::Request::getLookaheadConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request13getLoraConfigEv", "tensorrt_llm::executor::Request::getLoraConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getMaxTokensEv", "tensorrt_llm::executor::Request::getMaxTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getMropeConfigEv", "tensorrt_llm::executor::Request::getMropeConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getMultimodalEmbeddingEv", "tensorrt_llm::executor::Request::getMultimodalEmbedding"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request15getOutputConfigEv", "tensorrt_llm::executor::Request::getOutputConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request8getPadIdEv", "tensorrt_llm::executor::Request::getPadId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getPositionIdsEv", "tensorrt_llm::executor::Request::getPositionIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request11getPriorityEv", "tensorrt_llm::executor::Request::getPriority"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request21getPromptTuningConfigEv", "tensorrt_llm::executor::Request::getPromptTuningConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request14getRequestTypeEv", "tensorrt_llm::executor::Request::getRequestType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request27getReturnAllGeneratedTokensEv", "tensorrt_llm::executor::Request::getReturnAllGeneratedTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request17getSamplingConfigEv", "tensorrt_llm::executor::Request::getSamplingConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request22getSkipCrossAttnBlocksEv", "tensorrt_llm::executor::Request::getSkipCrossAttnBlocks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getStopWordsEv", "tensorrt_llm::executor::Request::getStopWords"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor7Request12getStreamingEv", "tensorrt_llm::executor::Request::getStreaming"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request25kBatchedPostProcessorNameE", "tensorrt_llm::executor::Request::kBatchedPostProcessorName"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request16kDefaultPriorityE", "tensorrt_llm::executor::Request::kDefaultPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request31kDynamicPostProcessorNamePrefixE", "tensorrt_llm::executor::Request::kDynamicPostProcessorNamePrefix"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor7Request5mImplE", "tensorrt_llm::executor::Request::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", "tensorrt_llm::executor::Request::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", "tensorrt_llm::executor::Request::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERK7Request", "tensorrt_llm::executor::Request::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7RequestaSERR7Request", "tensorrt_llm::executor::Request::operator=::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", "tensorrt_llm::executor::Request::setAllottedTimeMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request17setAllottedTimeMsE16MillisecondsType", "tensorrt_llm::executor::Request::setAllottedTimeMs::allottedTimeMs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setBadWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setBadWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setBadWords::badWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", "tensorrt_llm::executor::Request::setClientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setClientIdE6IdType", "tensorrt_llm::executor::Request::setClientId::clientId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", "tensorrt_llm::executor::Request::setContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setContextPhaseParamsE18ContextPhaseParams", "tensorrt_llm::executor::Request::setContextPhaseParams::contextPhaseParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", "tensorrt_llm::executor::Request::setCrossAttentionMask"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setCrossAttentionMaskE6Tensor", "tensorrt_llm::executor::Request::setCrossAttentionMask::crossAttentionMask"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::Request::setEagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setEagleConfigERKNSt8optionalI11EagleConfigEE", "tensorrt_llm::executor::Request::setEagleConfig::eagleConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", "tensorrt_llm::executor::Request::setEmbeddingBias"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request16setEmbeddingBiasERK6Tensor", "tensorrt_llm::executor::Request::setEmbeddingBias::embeddingBias"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", "tensorrt_llm::executor::Request::setEncoderInputFeatures"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputFeaturesE6Tensor", "tensorrt_llm::executor::Request::setEncoderInputFeatures::encoderInputFeatures"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", "tensorrt_llm::executor::Request::setEncoderInputTokenIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setEncoderInputTokenIdsERK9VecTokens", "tensorrt_llm::executor::Request::setEncoderInputTokenIds::encoderInputTokenIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", "tensorrt_llm::executor::Request::setEncoderOutputLength"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setEncoderOutputLengthE10SizeType32", "tensorrt_llm::executor::Request::setEncoderOutputLength::encoderOutputLength"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", "tensorrt_llm::executor::Request::setEndId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request8setEndIdE10SizeType32", "tensorrt_llm::executor::Request::setEndId::endId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Request::setExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request28setExternalDraftTokensConfigERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Request::setExternalDraftTokensConfig::externalDraftTokensConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", "tensorrt_llm::executor::Request::setGuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request23setGuidedDecodingParamsERK20GuidedDecodingParams", "tensorrt_llm::executor::Request::setGuidedDecodingParams::guidedDecodingParams"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Request::setKvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request25setKvCacheRetentionConfigERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Request::setKvCacheRetentionConfig::kvCacheRetentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", "tensorrt_llm::executor::Request::setLanguageAdapterUid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setLanguageAdapterUidE10SizeType32", "tensorrt_llm::executor::Request::setLanguageAdapterUid::languageAdapterUid"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", "tensorrt_llm::executor::Request::setLogitsPostProcessor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setLogitsPostProcessorERKNSt8optionalI19LogitsPostProcessorEE", "tensorrt_llm::executor::Request::setLogitsPostProcessor::logitsPostProcessor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", "tensorrt_llm::executor::Request::setLogitsPostProcessorName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request26setLogitsPostProcessorNameERKNSt6stringE", "tensorrt_llm::executor::Request::setLogitsPostProcessorName::logitsPostProcessorName"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Request::setLookaheadConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request18setLookaheadConfigERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Request::setLookaheadConfig::lookaheadConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", "tensorrt_llm::executor::Request::setLoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request13setLoraConfigERK10LoraConfig", "tensorrt_llm::executor::Request::setLoraConfig::loraConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", "tensorrt_llm::executor::Request::setMropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setMropeConfigERK11MropeConfig", "tensorrt_llm::executor::Request::setMropeConfig::mRopeConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", "tensorrt_llm::executor::Request::setMultimodalEmbedding"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setMultimodalEmbeddingERK6Tensor", "tensorrt_llm::executor::Request::setMultimodalEmbedding::multimodalEmbedding"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", "tensorrt_llm::executor::Request::setOutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request15setOutputConfigERK12OutputConfig", "tensorrt_llm::executor::Request::setOutputConfig::outputConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", "tensorrt_llm::executor::Request::setPadId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request8setPadIdE10SizeType32", "tensorrt_llm::executor::Request::setPadId::padId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::Request::setPositionIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setPositionIdsERKNSt6vectorI10SizeType32EE", "tensorrt_llm::executor::Request::setPositionIds::positionIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", "tensorrt_llm::executor::Request::setPriority"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request11setPriorityE12PriorityType", "tensorrt_llm::executor::Request::setPriority::priority"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", "tensorrt_llm::executor::Request::setPromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request21setPromptTuningConfigERK18PromptTuningConfig", "tensorrt_llm::executor::Request::setPromptTuningConfig::pTuningConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", "tensorrt_llm::executor::Request::setRequestType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request14setRequestTypeERK11RequestType", "tensorrt_llm::executor::Request::setRequestType::requestType"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", "tensorrt_llm::executor::Request::setReturnAllGeneratedTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request27setReturnAllGeneratedTokensEb", "tensorrt_llm::executor::Request::setReturnAllGeneratedTokens::returnAllGeneratedTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", "tensorrt_llm::executor::Request::setSamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request17setSamplingConfigERK14SamplingConfig", "tensorrt_llm::executor::Request::setSamplingConfig::config"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", "tensorrt_llm::executor::Request::setSkipCrossAttnBlocks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request22setSkipCrossAttnBlocksE6Tensor", "tensorrt_llm::executor::Request::setSkipCrossAttnBlocks::skipCrossAttnBlocks"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setStopWords"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStopWordsERKNSt4listI9VecTokensEE", "tensorrt_llm::executor::Request::setStopWords::stopWords"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", "tensorrt_llm::executor::Request::setStreaming"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor7Request12setStreamingEb", "tensorrt_llm::executor::Request::setStreaming::streaming"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7RequestD0Ev", "tensorrt_llm::executor::Request::~Request"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetricsE", "tensorrt_llm::executor::RequestPerfMetrics"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics14kvCacheHitRateE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::kvCacheHitRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numMissedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numMissedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics21numNewAllocatedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numNewAllocatedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics15numReusedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numReusedBlocks"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14KvCacheMetrics23numTotalAllocatedBlocksE", "tensorrt_llm::executor::RequestPerfMetrics::KvCacheMetrics::numTotalAllocatedBlocks"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics14acceptanceRateE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::acceptanceRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics24totalAcceptedDraftTokensE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::totalAcceptedDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics26SpeculativeDecodingMetrics16totalDraftTokensE", "tensorrt_llm::executor::RequestPerfMetrics::SpeculativeDecodingMetrics::totalDraftTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9TimePointE", "tensorrt_llm::executor::RequestPerfMetrics::TimePoint"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11arrivalTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::arrivalTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18firstScheduledTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::firstScheduledTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics14firstTokenTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::firstTokenTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics11kvCacheSizeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics18kvCacheTransferEndE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheTransferEnd"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics20kvCacheTransferStartE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::kvCacheTransferStart"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13TimingMetrics13lastTokenTimeE", "tensorrt_llm::executor::RequestPerfMetrics::TimingMetrics::lastTokenTime"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics9firstIterE", "tensorrt_llm::executor::RequestPerfMetrics::firstIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics4iterE", "tensorrt_llm::executor::RequestPerfMetrics::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics14kvCacheMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::kvCacheMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics8lastIterE", "tensorrt_llm::executor::RequestPerfMetrics::lastIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics19speculativeDecodingE", "tensorrt_llm::executor::RequestPerfMetrics::speculativeDecoding"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor18RequestPerfMetrics13timingMetricsE", "tensorrt_llm::executor::RequestPerfMetrics::timingMetrics"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor12RequestStageE", "tensorrt_llm::executor::RequestStage"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kCONTEXT_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kCONTEXT_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kENCODER_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kENCODER_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage20kGENERATION_COMPLETEE", "tensorrt_llm::executor::RequestStage::kGENERATION_COMPLETE"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage23kGENERATION_IN_PROGRESSE", "tensorrt_llm::executor::RequestStage::kGENERATION_IN_PROGRESS"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor12RequestStage7kQUEUEDE", "tensorrt_llm::executor::RequestStage::kQUEUED"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor12RequestStatsE", "tensorrt_llm::executor::RequestStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats24allocNewBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::allocNewBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats26allocTotalBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::allocTotalBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats26avgNumDecodedTokensPerIterE", "tensorrt_llm::executor::RequestStats::avgNumDecodedTokensPerIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22contextPrefillPositionE", "tensorrt_llm::executor::RequestStats::contextPrefillPosition"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats15disServingStatsE", "tensorrt_llm::executor::RequestStats::disServingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats2idE", "tensorrt_llm::executor::RequestStats::id"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats24kvCacheHitRatePerRequestE", "tensorrt_llm::executor::RequestStats::kvCacheHitRatePerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22missedBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::missedBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats18numGeneratedTokensE", "tensorrt_llm::executor::RequestStats::numGeneratedTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats6pausedE", "tensorrt_llm::executor::RequestStats::paused"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats22reusedBlocksPerRequestE", "tensorrt_llm::executor::RequestStats::reusedBlocksPerRequest"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats9scheduledE", "tensorrt_llm::executor::RequestStats::scheduled"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor12RequestStats5stageE", "tensorrt_llm::executor::RequestStats::stage"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIterationE", "tensorrt_llm::executor::RequestStatsPerIteration"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration4iterE", "tensorrt_llm::executor::RequestStatsPerIteration::iter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor24RequestStatsPerIteration12requestStatsE", "tensorrt_llm::executor::RequestStatsPerIteration::requestStats"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor11RequestTypeE", "tensorrt_llm::executor::RequestType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType35REQUEST_TYPE_CONTEXT_AND_GENERATIONE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_CONTEXT_AND_GENERATION"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType25REQUEST_TYPE_CONTEXT_ONLYE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_CONTEXT_ONLY"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor11RequestType28REQUEST_TYPE_GENERATION_ONLYE", "tensorrt_llm::executor::RequestType::REQUEST_TYPE_GENERATION_ONLY"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8ResponseE", "tensorrt_llm::executor::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", "tensorrt_llm::executor::Response::Response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", "tensorrt_llm::executor::Response::Response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::Result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::clientId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::errorMsg"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERK8Response", "tensorrt_llm::executor::Response::Response::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseERR8Response", "tensorrt_llm::executor::Response::Response::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdType6ResultNSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::requestId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8Response8ResponseE6IdTypeNSt6stringENSt8optionalI6IdTypeEE", "tensorrt_llm::executor::Response::Response::requestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response11getClientIdEv", "tensorrt_llm::executor::Response::getClientId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response11getErrorMsgEv", "tensorrt_llm::executor::Response::getErrorMsg"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response12getRequestIdEv", "tensorrt_llm::executor::Response::getRequestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response9getResultEv", "tensorrt_llm::executor::Response::getResult"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8Response8hasErrorEv", "tensorrt_llm::executor::Response::hasError"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8Response5mImplE", "tensorrt_llm::executor::Response::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", "tensorrt_llm::executor::Response::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", "tensorrt_llm::executor::Response::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERK8Response", "tensorrt_llm::executor::Response::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8ResponseaSERR8Response", "tensorrt_llm::executor::Response::operator=::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8ResponseD0Ev", "tensorrt_llm::executor::Response::~Response"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor6ResultE", "tensorrt_llm::executor::Result"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result17additionalOutputsE", "tensorrt_llm::executor::Result::additionalOutputs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13contextLogitsE", "tensorrt_llm::executor::Result::contextLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result18contextPhaseParamsE", "tensorrt_llm::executor::Result::contextPhaseParams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result11cumLogProbsE", "tensorrt_llm::executor::Result::cumLogProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result12decodingIterE", "tensorrt_llm::executor::Result::decodingIter"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13encoderOutputE", "tensorrt_llm::executor::Result::encoderOutput"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13finishReasonsE", "tensorrt_llm::executor::Result::finishReasons"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result16generationLogitsE", "tensorrt_llm::executor::Result::generationLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result7isFinalE", "tensorrt_llm::executor::Result::isFinal"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result15isSequenceFinalE", "tensorrt_llm::executor::Result::isSequenceFinal"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result8logProbsE", "tensorrt_llm::executor::Result::logProbs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result14outputTokenIdsE", "tensorrt_llm::executor::Result::outputTokenIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result18requestPerfMetricsE", "tensorrt_llm::executor::Result::requestPerfMetrics"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result13sequenceIndexE", "tensorrt_llm::executor::Result::sequenceIndex"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Result21specDecFastLogitsInfoE", "tensorrt_llm::executor::Result::specDecFastLogitsInfo"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor17RetentionPriorityE", "tensorrt_llm::executor::RetentionPriority"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDurationE", "tensorrt_llm::executor::RetentionPriorityAndDuration"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration::durationMs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration28RetentionPriorityAndDurationERKNSt8optionalI17RetentionPriorityEERKNSt8optionalINSt6chrono12millisecondsEEE", "tensorrt_llm::executor::RetentionPriorityAndDuration::RetentionPriorityAndDuration::retentionPriority"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration10durationMsE", "tensorrt_llm::executor::RetentionPriorityAndDuration::durationMs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor28RetentionPriorityAndDuration17retentionPriorityE", "tensorrt_llm::executor::RetentionPriorityAndDuration::retentionPriority"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfigE", "tensorrt_llm::executor::SamplingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::beamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::earlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::frequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::lengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::minP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::minTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::noRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::numReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::presencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::repetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::seed"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::temperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14SamplingConfigE10SizeType32RKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI11TokenIdTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI14RandomSeedTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI9FloatTypeEERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI10SizeType32EERKNSt8optionalI9FloatTypeEERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::SamplingConfig::topPResetIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkBeamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig28checkBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkBeamSearchDiversityRate::beamSearchDiversityRate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidth::beamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19checkBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEEK10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkBeamWidthArray::beamWidthArray"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkEarlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkEarlyStopping::earlyStopping"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkLengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18checkLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkLengthPenalty::lengthPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkMinP::minP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkMinTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkMinTokens::minTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::checkNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences::beamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig23checkNumReturnSequencesERKNSt8optionalI10SizeType32EE10SizeType32", "tensorrt_llm::executor::SamplingConfig::checkNumReturnSequences::numReturnSequences"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig22checkRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkRepetitionPenalty::repetitionpenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16checkTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTemperature::temperature"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopKERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopK::topK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig9checkTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopP::topP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14checkTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPDecay::topPDecay"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12checkTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPMin::topPMin"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPResetIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17checkTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::checkTopPResetIds::topPResetIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig26getBeamSearchDiversityRateEv", "tensorrt_llm::executor::SamplingConfig::getBeamSearchDiversityRate"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getBeamWidthEv", "tensorrt_llm::executor::SamplingConfig::getBeamWidth"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getBeamWidthArrayEv", "tensorrt_llm::executor::SamplingConfig::getBeamWidthArray"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getEarlyStoppingEv", "tensorrt_llm::executor::SamplingConfig::getEarlyStopping"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig19getFrequencyPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getFrequencyPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig16getLengthPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getLengthPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getMinPEv", "tensorrt_llm::executor::SamplingConfig::getMinP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getMinTokensEv", "tensorrt_llm::executor::SamplingConfig::getMinTokens"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getNoRepeatNgramSizeEv", "tensorrt_llm::executor::SamplingConfig::getNoRepeatNgramSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig17getNumReturnBeamsEv", "tensorrt_llm::executor::SamplingConfig::getNumReturnBeams"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig21getNumReturnSequencesEv", "tensorrt_llm::executor::SamplingConfig::getNumReturnSequences"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig18getPresencePenaltyEv", "tensorrt_llm::executor::SamplingConfig::getPresencePenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig20getRepetitionPenaltyEv", "tensorrt_llm::executor::SamplingConfig::getRepetitionPenalty"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getSeedEv", "tensorrt_llm::executor::SamplingConfig::getSeed"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig14getTemperatureEv", "tensorrt_llm::executor::SamplingConfig::getTemperature"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopKEv", "tensorrt_llm::executor::SamplingConfig::getTopK"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig7getTopPEv", "tensorrt_llm::executor::SamplingConfig::getTopP"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig12getTopPDecayEv", "tensorrt_llm::executor::SamplingConfig::getTopPDecay"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig10getTopPMinEv", "tensorrt_llm::executor::SamplingConfig::getTopPMin"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfig15getTopPResetIdsEv", "tensorrt_llm::executor::SamplingConfig::getTopPResetIds"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig24mBeamSearchDiversityRateE", "tensorrt_llm::executor::SamplingConfig::mBeamSearchDiversityRate"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mBeamWidthE", "tensorrt_llm::executor::SamplingConfig::mBeamWidth"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mBeamWidthArrayE", "tensorrt_llm::executor::SamplingConfig::mBeamWidthArray"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mEarlyStoppingE", "tensorrt_llm::executor::SamplingConfig::mEarlyStopping"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17mFrequencyPenaltyE", "tensorrt_llm::executor::SamplingConfig::mFrequencyPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14mLengthPenaltyE", "tensorrt_llm::executor::SamplingConfig::mLengthPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mMinPE", "tensorrt_llm::executor::SamplingConfig::mMinP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mMinTokensE", "tensorrt_llm::executor::SamplingConfig::mMinTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mNoRepeatNgramSizeE", "tensorrt_llm::executor::SamplingConfig::mNoRepeatNgramSize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15mNumReturnBeamsE", "tensorrt_llm::executor::SamplingConfig::mNumReturnBeams"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19mNumReturnSequencesE", "tensorrt_llm::executor::SamplingConfig::mNumReturnSequences"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16mPresencePenaltyE", "tensorrt_llm::executor::SamplingConfig::mPresencePenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18mRepetitionPenaltyE", "tensorrt_llm::executor::SamplingConfig::mRepetitionPenalty"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mSeedE", "tensorrt_llm::executor::SamplingConfig::mSeed"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12mTemperatureE", "tensorrt_llm::executor::SamplingConfig::mTemperature"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopKE", "tensorrt_llm::executor::SamplingConfig::mTopK"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig5mTopPE", "tensorrt_llm::executor::SamplingConfig::mTopP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10mTopPDecayE", "tensorrt_llm::executor::SamplingConfig::mTopPDecay"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig8mTopPMinE", "tensorrt_llm::executor::SamplingConfig::mTopPMin"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig13mTopPResetIdsE", "tensorrt_llm::executor::SamplingConfig::mTopPResetIds"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::executor::SamplingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::executor::SamplingConfig::operator==::other"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setBeamSearchDiversityRate"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig26setBeamSearchDiversityRateERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setBeamSearchDiversityRate::beamSearchDiversityRate"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::setBeamWidth"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setBeamWidthE10SizeType32", "tensorrt_llm::executor::SamplingConfig::setBeamWidth::beamWidth"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::setBeamWidthArray"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig17setBeamWidthArrayERKNSt8optionalINSt6vectorI10SizeType32EEEE", "tensorrt_llm::executor::SamplingConfig::setBeamWidthArray::beamWidthArray"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setEarlyStopping"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setEarlyStoppingERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setEarlyStopping::earlyStopping"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setFrequencyPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig19setFrequencyPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setFrequencyPenalty::frequencyPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setLengthPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig16setLengthPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setLengthPenalty::lengthPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setMinP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setMinPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setMinP::minP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setMinTokens"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setMinTokensERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setMinTokens::minTokens"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNoRepeatNgramSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setNoRepeatNgramSizeERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNoRepeatNgramSize::noRepeatNgramSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNumReturnSequences"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig21setNumReturnSequencesERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setNumReturnSequences::numReturnSequences"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setPresencePenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig18setPresencePenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setPresencePenalty::presencePenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setRepetitionPenalty"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20setRepetitionPenaltyERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setRepetitionPenalty::repetitionPenalty"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", "tensorrt_llm::executor::SamplingConfig::setSeed"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setSeedERKNSt8optionalI14RandomSeedTypeEE", "tensorrt_llm::executor::SamplingConfig::setSeed::seed"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTemperature"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig14setTemperatureERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTemperature::temperature"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setTopK"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopKERKNSt8optionalI10SizeType32EE", "tensorrt_llm::executor::SamplingConfig::setTopK::topK"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig7setTopPERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopP::topP"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPDecay"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig12setTopPDecayERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPDecay::topPDecay"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPMin"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig10setTopPMinERKNSt8optionalI9FloatTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPMin::topPMin"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPResetIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig15setTopPResetIdsERKNSt8optionalI11TokenIdTypeEE", "tensorrt_llm::executor::SamplingConfig::setTopPResetIds::topPResetIds"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor14SamplingConfig20updateNumReturnBeamsEv", "tensorrt_llm::executor::SamplingConfig::updateNumReturnBeams"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfigE", "tensorrt_llm::executor::SchedulerConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::capacitySchedulerPolicy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::contextChunkingPolicy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig15SchedulerConfigE23CapacitySchedulerPolicyNSt8optionalI21ContextChunkingPolicyEENSt8optionalI18DynamicBatchConfigEE", "tensorrt_llm::executor::SchedulerConfig::SchedulerConfig::dynamicBatchConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig26getCapacitySchedulerPolicyEv", "tensorrt_llm::executor::SchedulerConfig::getCapacitySchedulerPolicy"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig24getContextChunkingPolicyEv", "tensorrt_llm::executor::SchedulerConfig::getContextChunkingPolicy"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfig21getDynamicBatchConfigEv", "tensorrt_llm::executor::SchedulerConfig::getDynamicBatchConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig24mCapacitySchedulerPolicyE", "tensorrt_llm::executor::SchedulerConfig::mCapacitySchedulerPolicy"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig22mContextChunkingPolicyE", "tensorrt_llm::executor::SchedulerConfig::mContextChunkingPolicy"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15SchedulerConfig19mDynamicBatchConfigE", "tensorrt_llm::executor::SchedulerConfig::mDynamicBatchConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", "tensorrt_llm::executor::SchedulerConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor15SchedulerConfigeqERK15SchedulerConfig", "tensorrt_llm::executor::SchedulerConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor13SerializationE", "tensorrt_llm::executor::Serialization"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeAdditionalModelOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalModelOutput::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization27deserializeAdditionalOutputERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAdditionalOutput::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeAgentStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAgentState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeAgentStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeAgentState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeBool"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization15deserializeBoolERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeBool::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeCacheStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeCacheTransceiverConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCacheTransceiverConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeCommStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeCommState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeContextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeContextPhaseParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeContextPhaseParams::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeDataTransceiverStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDataTransceiverState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDebugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeDebugConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDebugConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeDecodingModeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDecodingMode::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDisServingRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeDisServingRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDisServingRequestStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeDynamicBatchConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeDynamicBatchConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeEagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeEagleConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeEagleConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExecutorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeExecutorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExecutorConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExtendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization40deserializeExtendedRuntimePerfKnobConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExtendedRuntimePerfKnobConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExternalDraftTokensConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeExternalDraftTokensConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeExternalDraftTokensConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization31deserializeGuidedDecodingParamsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeGuidedDecodingParams::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeInflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeInflightBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeInflightBatchingStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStats"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeIterationStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStats::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeIterationStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeIterationStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeIterationStatsVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeIterationStatsVec::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization24deserializeKvCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization33deserializeKvCacheRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheRetentionConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeKvCacheStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeKvCacheStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization34deserializeLookaheadDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLookaheadDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLoraConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization21deserializeLoraConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeLoraConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeModelType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeModelTypeERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeModelType::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeMropeConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeMropeConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeMropeConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOrchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeOrchestratorConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOrchestratorConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOutputConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeOutputConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeOutputConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeParallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeParallelConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeParallelConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePeftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializePeftCacheConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePeftCacheConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePromptTuningConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializePromptTuningConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializePromptTuningConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization18deserializeRequestERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequest::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestPerfMetrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization29deserializeRequestPerfMetricsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestPerfMetrics::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStageERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStage::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization23deserializeRequestStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration::buffer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization35deserializeRequestStatsPerIterationERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIteration::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIterationVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization38deserializeRequestStatsPerIterationVecERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeRequestStatsPerIterationVec::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResponse"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization19deserializeResponseERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResponse::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeResponsesERNSt6vectorIcEE", "tensorrt_llm::executor::Serialization::deserializeResponses::buffer"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResult"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeResultERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeResult::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSamplingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization25deserializeSamplingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSamplingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSchedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization26deserializeSchedulerConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSchedulerConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSocketState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization22deserializeSocketStateERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSocketState::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecFastLogitsInfo"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization32deserializeSpecDecFastLogitsInfoERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecFastLogitsInfo::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeSpecDecodingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecodingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization28deserializeSpecDecodingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpecDecodingStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpeculativeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeSpeculativeDecodingConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeSpeculativeDecodingConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeStaticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization30deserializeStaticBatchingStatsERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeStaticBatchingStats::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeString"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeStringERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeString::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization17deserializeTensorERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTensor::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTimePoint"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization20deserializeTimePointERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTimePoint::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization36deserializeTokenRangeRetentionConfigERNSt7istreamE", "tensorrt_llm::executor::Serialization::deserializeTokenRangeRetentionConfig::is"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK17SpecDecodingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10AgentStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", "tensorrt_llm::executor::Serialization::serialize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::additionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::additionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serialize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::dynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::inflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::info"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStats", "tensorrt_llm::executor::Serialization::serialize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI14IterationStatsEE", "tensorrt_llm::executor::Serialization::serialize::iterStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::kvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::metrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK10LoraConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11DebugConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11EagleConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK11MropeConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12DecodingModeRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12KvCacheStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12OutputConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK13KvCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14DecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ExecutorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14IterationStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14SamplingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK16AdditionalOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK17SpecDecodingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18ContextPhaseParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18DynamicBatchConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18OrchestratorConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18PromptTuningConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK18RequestPerfMetricsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20DataTransceiverStateRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK20GuidedDecodingParamsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21AdditionalModelOutputRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK21InflightBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22CacheTransceiverConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22KvCacheRetentionConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK23LookaheadDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25ExternalDraftTokensConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK29ExtendedRuntimePerfKnobConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK33SpeculativeDecodingFastLogitsInfoRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10AgentStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK14ParallelConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15PeftCacheConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK7RequestRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStageRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::requestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI24RequestStatsPerIterationEE", "tensorrt_llm::executor::Serialization::serialize::requestStatsVec"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK8ResponseRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKNSt6vectorI8ResponseEE", "tensorrt_llm::executor::Serialization::serialize::responses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6ResultRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK15SchedulerConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK25SpeculativeDecodingConfigRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK17SpecDecodingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::specDecStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK12RequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK24RequestStatsPerIterationRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10AgentStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache10CacheStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache11SocketStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN8kv_cache9CommStateERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK19StaticBatchingStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::staticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK22DisServingRequestStatsRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::stats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERK6TensorRNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tokenRangeRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization9serializeERKN18RequestPerfMetrics9TimePointERNSt7ostreamE", "tensorrt_llm::executor::Serialization::serialize::tp"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK17SpecDecodingStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN18RequestPerfMetrics9TimePointE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10AgentStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", "tensorrt_llm::executor::Serialization::serializedSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21AdditionalModelOutput", "tensorrt_llm::executor::Serialization::serializedSize::additionalModelOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK16AdditionalOutput", "tensorrt_llm::executor::Serialization::serializedSize::additionalOutput"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22CacheTransceiverConfig", "tensorrt_llm::executor::Serialization::serializedSize::cacheTransceiverConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK10LoraConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11MropeConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12OutputConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14SamplingConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18PromptTuningConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25ExternalDraftTokensConfig", "tensorrt_llm::executor::Serialization::serializedSize::config"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18ContextPhaseParams", "tensorrt_llm::executor::Serialization::serializedSize::contextPhaseParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20DataTransceiverState", "tensorrt_llm::executor::Serialization::serializedSize::dataTransceiverState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11DebugConfig", "tensorrt_llm::executor::Serialization::serializedSize::debugConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14DecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::decodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12DecodingMode", "tensorrt_llm::executor::Serialization::serializedSize::decodingMode"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22DisServingRequestStats", "tensorrt_llm::executor::Serialization::serializedSize::disServingRequestStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18DynamicBatchConfig", "tensorrt_llm::executor::Serialization::serializedSize::dynamicBatchConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK11EagleConfig", "tensorrt_llm::executor::Serialization::serializedSize::eagleConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ExecutorConfig", "tensorrt_llm::executor::Serialization::serializedSize::executorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK29ExtendedRuntimePerfKnobConfig", "tensorrt_llm::executor::Serialization::serializedSize::extendedRuntimePerfKnobConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::guidedDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK20GuidedDecodingParams", "tensorrt_llm::executor::Serialization::serializedSize::guidedDecodingParams"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK21InflightBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize::inflightBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK33SpeculativeDecodingFastLogitsInfo", "tensorrt_llm::executor::Serialization::serializedSize::info"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14IterationStats", "tensorrt_llm::executor::Serialization::serializedSize::iterStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK13KvCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK22KvCacheRetentionConfig", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheRetentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12KvCacheStats", "tensorrt_llm::executor::Serialization::serializedSize::kvCacheStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK23LookaheadDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::lookaheadDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18RequestPerfMetrics", "tensorrt_llm::executor::Serialization::serializedSize::metrics"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK18OrchestratorConfig", "tensorrt_llm::executor::Serialization::serializedSize::orchestratorConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK14ParallelConfig", "tensorrt_llm::executor::Serialization::serializedSize::parallelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15PeftCacheConfig", "tensorrt_llm::executor::Serialization::serializedSize::peftCacheConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK7Request", "tensorrt_llm::executor::Serialization::serializedSize::request"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStage", "tensorrt_llm::executor::Serialization::serializedSize::requestStage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK8Response", "tensorrt_llm::executor::Serialization::serializedSize::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Result", "tensorrt_llm::executor::Serialization::serializedSize::result"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK15SchedulerConfig", "tensorrt_llm::executor::Serialization::serializedSize::schedulerConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::Serialization::serializedSize::specDecConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK17SpecDecodingStats", "tensorrt_llm::executor::Serialization::serializedSize::specDecStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK12RequestStats", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK24RequestStatsPerIteration", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10AgentStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache11SocketStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN8kv_cache9CommStateE", "tensorrt_llm::executor::Serialization::serializedSize::state"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK19StaticBatchingStats", "tensorrt_llm::executor::Serialization::serializedSize::staticBatchingStats"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERK6Tensor", "tensorrt_llm::executor::Serialization::serializedSize::tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor13Serialization14serializedSizeERKN22KvCacheRetentionConfig25TokenRangeRetentionConfigE", "tensorrt_llm::executor::Serialization::serializedSize::tokenRangeRetentionConfig"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor5ShapeE", "tensorrt_llm::executor::Shape"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor5Shape4BaseE", "tensorrt_llm::executor::Shape::Base"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor5Shape9DimType64E", "tensorrt_llm::executor::Shape::DimType64"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", "tensorrt_llm::executor::Shape::Shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEv", "tensorrt_llm::executor::Shape::Shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeENSt16initializer_listI9DimType64EE", "tensorrt_llm::executor::Shape::Shape::dims"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor5Shape5ShapeEPK9DimType64N4Base9size_typeE", "tensorrt_llm::executor::Shape::Shape::size"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10SizeType32E", "tensorrt_llm::executor::SizeType32"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor10SizeType64E", "tensorrt_llm::executor::SizeType64"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStatsE", "tensorrt_llm::executor::SpecDecodingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats16acceptanceLengthE", "tensorrt_llm::executor::SpecDecodingStats::acceptanceLength"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats13draftOverheadE", "tensorrt_llm::executor::SpecDecodingStats::draftOverhead"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats13iterLatencyMSE", "tensorrt_llm::executor::SpecDecodingStats::iterLatencyMS"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats17numAcceptedTokensE", "tensorrt_llm::executor::SpecDecodingStats::numAcceptedTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats14numDraftTokensE", "tensorrt_llm::executor::SpecDecodingStats::numDraftTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor17SpecDecodingStats26numRequestsWithDraftTokensE", "tensorrt_llm::executor::SpecDecodingStats::numRequestsWithDraftTokens"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfigE", "tensorrt_llm::executor::SpeculativeDecodingConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", "tensorrt_llm::executor::SpeculativeDecodingConfig::SpeculativeDecodingConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig25SpeculativeDecodingConfigEb", "tensorrt_llm::executor::SpeculativeDecodingConfig::SpeculativeDecodingConfig::fastLogits"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor25SpeculativeDecodingConfig10fastLogitsE", "tensorrt_llm::executor::SpeculativeDecodingConfig::fastLogits"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::SpeculativeDecodingConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor25SpeculativeDecodingConfigeqERK25SpeculativeDecodingConfig", "tensorrt_llm::executor::SpeculativeDecodingConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfoE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo18draftParticipantIdE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::draftParticipantId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo14draftRequestIdE", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::draftRequestId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor33SpeculativeDecodingFastLogitsInfo8toTensorEv", "tensorrt_llm::executor::SpeculativeDecodingFastLogitsInfo::toTensor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStatsE", "tensorrt_llm::executor::StaticBatchingStats"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats13emptyGenSlotsE", "tensorrt_llm::executor::StaticBatchingStats::emptyGenSlots"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats18numContextRequestsE", "tensorrt_llm::executor::StaticBatchingStats::numContextRequests"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numCtxTokensE", "tensorrt_llm::executor::StaticBatchingStats::numCtxTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats12numGenTokensE", "tensorrt_llm::executor::StaticBatchingStats::numGenTokens"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor19StaticBatchingStats20numScheduledRequestsE", "tensorrt_llm::executor::StaticBatchingStats::numScheduledRequests"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9StreamPtrE", "tensorrt_llm::executor::StreamPtr"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor6TensorE", "tensorrt_llm::executor::Tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::CudaStreamPtr"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6Tensor4ImplE", "tensorrt_llm::executor::Tensor::Impl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", "tensorrt_llm::executor::Tensor::Tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorEv", "tensorrt_llm::executor::Tensor::Tensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERK6Tensor", "tensorrt_llm::executor::Tensor::Tensor::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorERR6Tensor", "tensorrt_llm::executor::Tensor::Tensor::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6TensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::Tensor::tensor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo::stream"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor6copyToENSt10shared_ptrI4ImplEE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::copyTo::tensor"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToCpu"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToCpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToCpu::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToGpu"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor9copyToGpuEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToGpu::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToManaged"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13copyToManagedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToManaged::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPinned"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor12copyToPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPinned::stream"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPooledPinned"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor18copyToPooledPinnedEN6Tensor13CudaStreamPtrE", "tensorrt_llm::executor::Tensor::copyToPooledPinned::stream"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3cpuE6Tensor5Shape", "tensorrt_llm::executor::Tensor::cpu::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3cpuE8DataType5Shape", "tensorrt_llm::executor::Tensor::cpu::shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::detail::ofITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::Tensor::detail::ofITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::Tensor::detail::toITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::Tensor::detail::toITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7getDataEv", "tensorrt_llm::executor::Tensor::getData"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor7getDataEv", "tensorrt_llm::executor::Tensor::getData"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor11getDataTypeEv", "tensorrt_llm::executor::Tensor::getDataType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor13getMemoryTypeEv", "tensorrt_llm::executor::Tensor::getMemoryType"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", "tensorrt_llm::executor::Tensor::getRuntimeType"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor14getRuntimeTypeE8DataTypev", "tensorrt_llm::executor::Tensor::getRuntimeType::T"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor8getShapeEv", "tensorrt_llm::executor::Tensor::getShape"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor7getSizeEv", "tensorrt_llm::executor::Tensor::getSize"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6Tensor14getSizeInBytesEv", "tensorrt_llm::executor::Tensor::getSizeInBytes"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::shape"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor3gpuE6Tensor13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::stream"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor3gpuE8DataType13CudaStreamPtr5Shape", "tensorrt_llm::executor::Tensor::gpu::stream"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7mTensorE", "tensorrt_llm::executor::Tensor::mTensor"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor7managedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::managed::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7managedE8DataType5Shape", "tensorrt_llm::executor::Tensor::managed::shape"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::T"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of::T"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorR1T", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor2ofE6TensorP1T5Shape", "tensorrt_llm::executor::Tensor::of::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor2ofE8DataTypePv5Shape", "tensorrt_llm::executor::Tensor::of::shape"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensorcvbEv", "tensorrt_llm::executor::Tensor::operator bool"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", "tensorrt_llm::executor::Tensor::operator!="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6TensorneERK6Tensor", "tensorrt_llm::executor::Tensor::operator!=::rhs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", "tensorrt_llm::executor::Tensor::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", "tensorrt_llm::executor::Tensor::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERK6Tensor", "tensorrt_llm::executor::Tensor::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6TensoraSERR6Tensor", "tensorrt_llm::executor::Tensor::operator=::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", "tensorrt_llm::executor::Tensor::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor6TensoreqERK6Tensor", "tensorrt_llm::executor::Tensor::operator==::rhs"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor6pinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pinned::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor6pinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pinned::shape"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::T"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::dataType"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor6Tensor12pooledPinnedE6Tensor5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::shape"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor12pooledPinnedE8DataType5Shape", "tensorrt_llm::executor::Tensor::pooledPinned::shape"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setFromERK6Tensor13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setFrom::stream"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setZero"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6Tensor7setZeroE13CudaStreamPtr", "tensorrt_llm::executor::Tensor::setZero::stream"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6TensorD0Ev", "tensorrt_llm::executor::Tensor::~Tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9TensorPtrE", "tensorrt_llm::executor::TensorPtr"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11TokenIdTypeE", "tensorrt_llm::executor::TokenIdType"], [0, 2, 1, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", "tensorrt_llm::executor::TypeTraits"], [0, 8, 1, "_CPPv4I0_bEN12tensorrt_llm8executor10TypeTraitsE", "tensorrt_llm::executor::TypeTraits::T"], [0, 2, 1, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", "tensorrt_llm::executor::TypeTraits<T*>"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor10TypeTraitsIP1TEE", "tensorrt_llm::executor::TypeTraits<T*>::T"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIP1TE5valueE", "tensorrt_llm::executor::TypeTraits<T*>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIbEE", "tensorrt_llm::executor::TypeTraits<bool>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIbE5valueE", "tensorrt_llm::executor::TypeTraits<bool>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsIfEE", "tensorrt_llm::executor::TypeTraits<float>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsIfE5valueE", "tensorrt_llm::executor::TypeTraits<float>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsI4halfEE", "tensorrt_llm::executor::TypeTraits<half>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsI4halfE5valueE", "tensorrt_llm::executor::TypeTraits<half>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int32_tEEE", "tensorrt_llm::executor::TypeTraits<std::int32_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int32_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int32_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7int64_tEEE", "tensorrt_llm::executor::TypeTraits<std::int64_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7int64_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int64_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt6int8_tEEE", "tensorrt_llm::executor::TypeTraits<std::int8_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt6int8_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::int8_t>::value"], [0, 2, 1, "_CPPv4IEN12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEEE", "tensorrt_llm::executor::TypeTraits<std::uint8_t>"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor10TypeTraitsINSt7uint8_tEE5valueE", "tensorrt_llm::executor::TypeTraits<std::uint8_t>::value"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor11VecLogProbsE", "tensorrt_llm::executor::VecLogProbs"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor16VecTokenExtraIdsE", "tensorrt_llm::executor::VecTokenExtraIds"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor9VecTokensE", "tensorrt_llm::executor::VecTokens"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6detailE", "tensorrt_llm::executor::detail"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor6detail9DimType64E", "tensorrt_llm::executor::detail::DimType64"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::detail::ofITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6detail9ofITensorENSt10shared_ptrIN7runtime7ITensorEEE", "tensorrt_llm::executor::detail::ofITensor::tensor"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::detail::toITensor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor6detail9toITensorERK6Tensor", "tensorrt_llm::executor::detail::toITensor::tensor"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executorE", "tensorrt_llm::executor::disagg_executor"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::ctxEnginePaths"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::ctxExecutorConfigs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::genEnginePaths"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::genExecutorConfigs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::hasContextAwaitThreads"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator26DisaggExecutorOrchestratorERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorINSt10filesystem4pathEEERKNSt6vectorIN8executor14ExecutorConfigEEERKNSt6vectorIN8executor14ExecutorConfigEEEbb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::DisaggExecutorOrchestrator::hasGenAwaitThreads"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses::contextIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator21awaitContextResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitContextResponses::timeout"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses::genIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator24awaitGenerationResponsesERKNSt8optionalINSt6chrono12millisecondsEEENSt8optionalIiEE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::awaitGenerationResponses::timeout"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator10canEnqueueEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::canEnqueue"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::batch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::requests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator14enqueueContextERKNSt6vectorIN5texec7RequestEEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueContext::selectContextId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::batch"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::globalRequestIds"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::requests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator17enqueueGenerationERKNSt6vectorIN5texec7RequestEEERKNSt6vectorI6IdTypeEENSt8optionalIiEEb", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::enqueueGeneration::selectGenIdx"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator19getContextExecutorsEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::getContextExecutors"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator15getGenExecutorsEv", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::getGenExecutors"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestrator5mImplE", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::mImpl"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor26DisaggExecutorOrchestratorD0Ev", "tensorrt_llm::executor::disagg_executor::DisaggExecutorOrchestrator::~DisaggExecutorOrchestrator"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdE", "tensorrt_llm::executor::disagg_executor::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::gid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::gid"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERKN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::response"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId14ResponseWithIdERRN12tensorrt_llm8executor8ResponseE6IdType", "tensorrt_llm::executor::disagg_executor::ResponseWithId::ResponseWithId::response"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId3gidE", "tensorrt_llm::executor::disagg_executor::ResponseWithId::gid"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator="], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERK14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator=::other"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdaSERR14ResponseWithId", "tensorrt_llm::executor::disagg_executor::ResponseWithId::operator=::other"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithId8responseE", "tensorrt_llm::executor::disagg_executor::ResponseWithId::response"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor15disagg_executor14ResponseWithIdD0Ev", "tensorrt_llm::executor::disagg_executor::ResponseWithId::~ResponseWithId"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cacheE", "tensorrt_llm::executor::kv_cache"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDescE", "tensorrt_llm::executor::kv_cache::AgentDesc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDesc9AgentDescENSt6stringE", "tensorrt_llm::executor::kv_cache::AgentDesc::AgentDesc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDesc9AgentDescENSt6stringE", "tensorrt_llm::executor::kv_cache::AgentDesc::AgentDesc::backendAgentDesc"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9AgentDesc19getBackendAgentDescEv", "tensorrt_llm::executor::kv_cache::AgentDesc::getBackendAgentDesc"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9AgentDesc17mBackendAgentDescE", "tensorrt_llm::executor::kv_cache::AgentDesc::mBackendAgentDesc"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentStateE", "tensorrt_llm::executor::kv_cache::AgentState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateENSt6stringENSt6stringE", "tensorrt_llm::executor::kv_cache::AgentState::AgentState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateEv", "tensorrt_llm::executor::kv_cache::AgentState::AgentState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateENSt6stringENSt6stringE", "tensorrt_llm::executor::kv_cache::AgentState::AgentState::agentName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10AgentStateENSt6stringENSt6stringE", "tensorrt_llm::executor::kv_cache::AgentState::AgentState::connectionInfo"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState10mAgentNameE", "tensorrt_llm::executor::kv_cache::AgentState::mAgentName"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10AgentState15mConnectionInfoE", "tensorrt_llm::executor::kv_cache::AgentState::mConnectionInfo"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10AgentStateeqERK10AgentState", "tensorrt_llm::executor::kv_cache::AgentState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10AgentStateeqERK10AgentState", "tensorrt_llm::executor::kv_cache::AgentState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10AgentState8toStringEv", "tensorrt_llm::executor::kv_cache::AgentState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfigE", "tensorrt_llm::executor::kv_cache::BaseAgentConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfig5mNameE", "tensorrt_llm::executor::kv_cache::BaseAgentConfig::mName"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15BaseAgentConfig13useProgThreadE", "tensorrt_llm::executor::kv_cache::BaseAgentConfig::useProgThread"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgentE", "tensorrt_llm::executor::kv_cache::BaseTransferAgent"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16checkRemoteDescsERKNSt6stringERK11MemoryDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::checkRemoteDescs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16checkRemoteDescsERKNSt6stringERK11MemoryDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::checkRemoteDescs::memoryDescs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16checkRemoteDescsERKNSt6stringERK11MemoryDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::checkRemoteDescs::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent18connectRemoteAgentERKNSt6stringERK18ConnectionInfoType", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::connectRemoteAgent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent18connectRemoteAgentERKNSt6stringERK18ConnectionInfoType", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::connectRemoteAgent::connectionInfo"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent18connectRemoteAgentERKNSt6stringERK18ConnectionInfoType", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::connectRemoteAgent::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16deregisterMemoryERK13RegisterDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::deregisterMemory"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent16deregisterMemoryERK13RegisterDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::deregisterMemory::descs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17getConnectionInfoEv", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::getConnectionInfo"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17getLocalAgentDescEv", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::getLocalAgentDesc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent23getNotifiedSyncMessagesEv", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::getNotifiedSyncMessages"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent21invalidateRemoteAgentERKNSt6stringE", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::invalidateRemoteAgent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent21invalidateRemoteAgentERKNSt6stringE", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::invalidateRemoteAgent::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent15loadRemoteAgentERKNSt6stringERK9AgentDesc", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::loadRemoteAgent"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent15loadRemoteAgentERKNSt6stringERK9AgentDesc", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::loadRemoteAgent::agentDesc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent15loadRemoteAgentERKNSt6stringERK9AgentDesc", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::loadRemoteAgent::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17notifySyncMessageERKNSt6stringERK11SyncMessage", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::notifySyncMessage"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17notifySyncMessageERKNSt6stringERK11SyncMessage", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::notifySyncMessage::name"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent17notifySyncMessageERKNSt6stringERK11SyncMessage", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::notifySyncMessage::syncMessage"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent14registerMemoryERK13RegisterDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::registerMemory"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent14registerMemoryERK13RegisterDescs", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::registerMemory::descs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent22submitTransferRequestsERK15TransferRequest", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::submitTransferRequests"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgent22submitTransferRequestsERK15TransferRequest", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::submitTransferRequests::request"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17BaseTransferAgentD0Ev", "tensorrt_llm::executor::kv_cache::BaseTransferAgent::~BaseTransferAgent"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfigE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig15AttentionConfigE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::AttentionConfig::kvFactor"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig14mAttentionTypeE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::mAttentionType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15AttentionConfig9mKvFactorE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionConfig::mKvFactor"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionTypeE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType8kDEFAULTE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType::kDEFAULT"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState13AttentionType4kMLAE", "tensorrt_llm::executor::kv_cache::CacheState::AttentionType::kMLA"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPrank"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPrank"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPsize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::DPsize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::attentionType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::dataType"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::enableAttentionDP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::enableAttentionDP"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::kvFactor"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::modelConfig"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbAttentionLayers"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbKvHeadPerLayer"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::nbKvHeads"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::pipelineParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::pipelineParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::sizePerHead"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::sizePerHead"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tensorParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tensorParallelism"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tokensPerBlock"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateENSt6vectorI10SizeType32EE10SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE13AttentionTypeibii", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::tokensPerBlock"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState10CacheStateE11ModelConfigRKN7runtime11WorldConfigEN8nvinfer18DataTypeE13AttentionTypei", "tensorrt_llm::executor::kv_cache::CacheState::CacheState::worldConfig"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig18mNbKvHeadsPerLayerE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mNbKvHeadsPerLayer"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig12mSizePerHeadE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mSizePerHead"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState11ModelConfig15mTokensPerBlockE", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::mTokensPerBlock"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11ModelConfigeqERK11ModelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ModelConfig::operator==::other"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPrankE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mDPrank"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig7mDPsizeE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mDPsize"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mEnableAttentionDPE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mEnableAttentionDP"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig20mPipelineParallelismE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mPipelineParallelism"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfig18mTensorParallelismE", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::mTensorParallelism"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14ParallelConfigeqERK14ParallelConfig", "tensorrt_llm::executor::kv_cache::CacheState::ParallelConfig::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState18getAttentionConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getAttentionConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState11getDataTypeEv", "tensorrt_llm::executor::kv_cache::CacheState::getDataType"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState14getModelConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getModelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState17getParallelConfigEv", "tensorrt_llm::executor::kv_cache::CacheState::getParallelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState16mAttentionConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mAttentionConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState9mDataTypeE", "tensorrt_llm::executor::kv_cache::CacheState::mDataType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState12mModelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mModelConfig"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10CacheState15mParallelConfigE", "tensorrt_llm::executor::kv_cache::CacheState::mParallelConfig"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheStateeqERKN8kv_cache10CacheStateE", "tensorrt_llm::executor::kv_cache::CacheState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10CacheState8toStringEv", "tensorrt_llm::executor::kv_cache::CacheState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommStateE", "tensorrt_llm::executor::kv_cache::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10AgentStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateEv", "tensorrt_llm::executor::kv_cache::CommState::CommState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10AgentStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::agentState"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState::ip"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt8uint16_tENSt6stringE", "tensorrt_llm::executor::kv_cache::CommState::CommState::port"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::ranks"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10AgentStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::selfIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI10SizeType32EEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::selfIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::selfIdx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState9CommStateENSt6vectorI11SocketStateEEi", "tensorrt_llm::executor::kv_cache::CommState::CommState::socketState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13getAgentStateEv", "tensorrt_llm::executor::kv_cache::CommState::getAgentState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState11getMpiStateEv", "tensorrt_llm::executor::kv_cache::CommState::getMpiState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10getSelfIdxEv", "tensorrt_llm::executor::kv_cache::CommState::getSelfIdx"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState14getSocketStateEv", "tensorrt_llm::executor::kv_cache::CommState::getSocketState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState12isAgentStateEv", "tensorrt_llm::executor::kv_cache::CommState::isAgentState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState10isMpiStateEv", "tensorrt_llm::executor::kv_cache::CommState::isMpiState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState13isSocketStateEv", "tensorrt_llm::executor::kv_cache::CommState::isSocketState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState8mSelfIdxE", "tensorrt_llm::executor::kv_cache::CommState::mSelfIdx"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache9CommState6mStateE", "tensorrt_llm::executor::kv_cache::CommState::mState"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", "tensorrt_llm::executor::kv_cache::CommState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommStateeqERK9CommState", "tensorrt_llm::executor::kv_cache::CommState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache9CommState8toStringEv", "tensorrt_llm::executor::kv_cache::CommState::toString"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionE", "tensorrt_llm::executor::kv_cache::Connection"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection12isThreadSafeEv", "tensorrt_llm::executor::kv_cache::Connection::isThreadSafe"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::ctx"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::data"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4recvERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::Connection::recv::size"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::ctx"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::data"], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10Connection4sendERK11DataContextPKv6size_t", "tensorrt_llm::executor::kv_cache::Connection::send::size"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10ConnectionD0Ev", "tensorrt_llm::executor::kv_cache::Connection::~Connection"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache18ConnectionInfoTypeE", "tensorrt_llm::executor::kv_cache::ConnectionInfoType"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerE", "tensorrt_llm::executor::kv_cache::ConnectionManager"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache17ConnectionManager12getCommStateEv", "tensorrt_llm::executor::kv_cache::ConnectionManager::getCommState"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", "tensorrt_llm::executor::kv_cache::ConnectionManager::getConnections"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager14getConnectionsERK9CommState", "tensorrt_llm::executor::kv_cache::ConnectionManager::getConnections::state"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::ctx"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::data"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManager11recvConnectERK11DataContextPv6size_t", "tensorrt_llm::executor::kv_cache::ConnectionManager::recvConnect::size"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache17ConnectionManagerD0Ev", "tensorrt_llm::executor::kv_cache::ConnectionManager::~ConnectionManager"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContextE", "tensorrt_llm::executor::kv_cache::DataContext"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", "tensorrt_llm::executor::kv_cache::DataContext::DataContext"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext11DataContextEi", "tensorrt_llm::executor::kv_cache::DataContext::DataContext::tag"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11DataContext6getTagEv", "tensorrt_llm::executor::kv_cache::DataContext::getTag"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11DataContext4mTagE", "tensorrt_llm::executor::kv_cache::DataContext::mTag"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderE", "tensorrt_llm::executor::kv_cache::DynLibLoader"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader12DynLibLoaderERK12DynLibLoader", "tensorrt_llm::executor::kv_cache::DynLibLoader::DynLibLoader"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader12DynLibLoaderEv", "tensorrt_llm::executor::kv_cache::DynLibLoader::DynLibLoader"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader5dlSymEPvPKc", "tensorrt_llm::executor::kv_cache::DynLibLoader::dlSym"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader5dlSymEPvPKc", "tensorrt_llm::executor::kv_cache::DynLibLoader::dlSym::handle"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader5dlSymEPvPKc", "tensorrt_llm::executor::kv_cache::DynLibLoader::dlSym::symbol"], [0, 3, 1, "_CPPv4I0EN12tensorrt_llm8executor8kv_cache12DynLibLoader18getFunctionPointerE9FunctionTRKNSt6stringERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getFunctionPointer"], [0, 8, 1, "_CPPv4I0EN12tensorrt_llm8executor8kv_cache12DynLibLoader18getFunctionPointerE9FunctionTRKNSt6stringERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getFunctionPointer::FunctionT"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor8kv_cache12DynLibLoader18getFunctionPointerE9FunctionTRKNSt6stringERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getFunctionPointer::funcName"], [0, 4, 1, "_CPPv4I0EN12tensorrt_llm8executor8kv_cache12DynLibLoader18getFunctionPointerE9FunctionTRKNSt6stringERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getFunctionPointer::libName"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9getHandleERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getHandle"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9getHandleERKNSt6stringE", "tensorrt_llm::executor::kv_cache::DynLibLoader::getHandle::name"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader11getInstanceEv", "tensorrt_llm::executor::kv_cache::DynLibLoader::getInstance"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9mDllMutexE", "tensorrt_llm::executor::kv_cache::DynLibLoader::mDllMutex"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoader9mHandlersE", "tensorrt_llm::executor::kv_cache::DynLibLoader::mHandlers"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderaSERK12DynLibLoader", "tensorrt_llm::executor::kv_cache::DynLibLoader::operator="], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache12DynLibLoaderD0Ev", "tensorrt_llm::executor::kv_cache::DynLibLoader::~DynLibLoader"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDescE", "tensorrt_llm::executor::kv_cache::MemoryDesc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescE9uintptr_t6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescEPv6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescERKNSt6vectorIcEE8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescE9uintptr_t6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::addr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescEPv6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::addr"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescE9uintptr_t6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::deviceId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescEPv6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::deviceId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescERKNSt6vectorIcEE8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::deviceId"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescE9uintptr_t6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::len"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescEPv6size_t8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::len"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc10MemoryDescERKNSt6vectorIcEE8uint32_t", "tensorrt_llm::executor::kv_cache::MemoryDesc::MemoryDesc::vec"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc11deserializeERNSt7istreamE", "tensorrt_llm::executor::kv_cache::MemoryDesc::deserialize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc11deserializeERNSt7istreamE", "tensorrt_llm::executor::kv_cache::MemoryDesc::deserialize::is"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc7getAddrEv", "tensorrt_llm::executor::kv_cache::MemoryDesc::getAddr"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc11getDeviceIdEv", "tensorrt_llm::executor::kv_cache::MemoryDesc::getDeviceId"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache10MemoryDesc6getLenEv", "tensorrt_llm::executor::kv_cache::MemoryDesc::getLen"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc5mAddrE", "tensorrt_llm::executor::kv_cache::MemoryDesc::mAddr"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9mDeviceIdE", "tensorrt_llm::executor::kv_cache::MemoryDesc::mDeviceId"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc4mLenE", "tensorrt_llm::executor::kv_cache::MemoryDesc::mLen"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9serializeERK10MemoryDescRNSt7ostreamE", "tensorrt_llm::executor::kv_cache::MemoryDesc::serialize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9serializeERK10MemoryDescRNSt7ostreamE", "tensorrt_llm::executor::kv_cache::MemoryDesc::serialize::memoryDesc"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc9serializeERK10MemoryDescRNSt7ostreamE", "tensorrt_llm::executor::kv_cache::MemoryDesc::serialize::os"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc14serializedSizeERK10MemoryDesc", "tensorrt_llm::executor::kv_cache::MemoryDesc::serializedSize"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryDesc14serializedSizeERK10MemoryDesc", "tensorrt_llm::executor::kv_cache::MemoryDesc::serializedSize::memoryDesc"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescsE", "tensorrt_llm::executor::kv_cache::MemoryDescs"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs11MemoryDescsE10MemoryTypeNSt6vectorI10MemoryDescEE", "tensorrt_llm::executor::kv_cache::MemoryDescs::MemoryDescs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs11MemoryDescsE10MemoryTypeNSt6vectorI10MemoryDescEE", "tensorrt_llm::executor::kv_cache::MemoryDescs::MemoryDescs::descs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs11MemoryDescsE10MemoryTypeNSt6vectorI10MemoryDescEE", "tensorrt_llm::executor::kv_cache::MemoryDescs::MemoryDescs::type"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11MemoryDescs8getDescsEv", "tensorrt_llm::executor::kv_cache::MemoryDescs::getDescs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11MemoryDescs7getTypeEv", "tensorrt_llm::executor::kv_cache::MemoryDescs::getType"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs6mDescsE", "tensorrt_llm::executor::kv_cache::MemoryDescs::mDescs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11MemoryDescs5mTypeE", "tensorrt_llm::executor::kv_cache::MemoryDescs::mType"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryTypeE", "tensorrt_llm::executor::kv_cache::MemoryType"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType4kBLKE", "tensorrt_llm::executor::kv_cache::MemoryType::kBLK"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kDRAME", "tensorrt_llm::executor::kv_cache::MemoryType::kDRAM"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kFILEE", "tensorrt_llm::executor::kv_cache::MemoryType::kFILE"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType4kOBJE", "tensorrt_llm::executor::kv_cache::MemoryType::kOBJ"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10MemoryType5kVRAME", "tensorrt_llm::executor::kv_cache::MemoryType::kVRAM"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiStateE", "tensorrt_llm::executor::kv_cache::MpiState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache8MpiState6mRanksE", "tensorrt_llm::executor::kv_cache::MpiState::mRanks"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", "tensorrt_llm::executor::kv_cache::MpiState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiStateeqERK8MpiState", "tensorrt_llm::executor::kv_cache::MpiState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache8MpiState8toStringEv", "tensorrt_llm::executor::kv_cache::MpiState::toString"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache13RegisterDescsE", "tensorrt_llm::executor::kv_cache::RegisterDescs"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketStateE", "tensorrt_llm::executor::kv_cache::SocketState"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState3mIpE", "tensorrt_llm::executor::kv_cache::SocketState::mIp"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SocketState5mPortE", "tensorrt_llm::executor::kv_cache::SocketState::mPort"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", "tensorrt_llm::executor::kv_cache::SocketState::operator=="], [0, 4, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketStateeqERK11SocketState", "tensorrt_llm::executor::kv_cache::SocketState::operator==::other"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache11SocketState8toStringEv", "tensorrt_llm::executor::kv_cache::SocketState::toString"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache11SyncMessageE", "tensorrt_llm::executor::kv_cache::SyncMessage"], [0, 1, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache13TransferDescsE", "tensorrt_llm::executor::kv_cache::TransferDescs"], [0, 6, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOpE", "tensorrt_llm::executor::kv_cache::TransferOp"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOp5kREADE", "tensorrt_llm::executor::kv_cache::TransferOp::kREAD"], [0, 7, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache10TransferOp6kWRITEE", "tensorrt_llm::executor::kv_cache::TransferOp::kWRITE"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequestE", "tensorrt_llm::executor::kv_cache::TransferRequest"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest::dstDescs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest::op"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest::remoteName"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest::srcDescs"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest15TransferRequestE10TransferOp13TransferDescs13TransferDescsRKNSt6stringENSt8optionalI11SyncMessageEE", "tensorrt_llm::executor::kv_cache::TransferRequest::TransferRequest::syncMessage"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest11getDstDescsEv", "tensorrt_llm::executor::kv_cache::TransferRequest::getDstDescs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest5getOpEv", "tensorrt_llm::executor::kv_cache::TransferRequest::getOp"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest13getRemoteNameEv", "tensorrt_llm::executor::kv_cache::TransferRequest::getRemoteName"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest11getSrcDescsEv", "tensorrt_llm::executor::kv_cache::TransferRequest::getSrcDescs"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache15TransferRequest14getSyncMessageEv", "tensorrt_llm::executor::kv_cache::TransferRequest::getSyncMessage"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest9mDstDescsE", "tensorrt_llm::executor::kv_cache::TransferRequest::mDstDescs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest3mOpE", "tensorrt_llm::executor::kv_cache::TransferRequest::mOp"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest11mRemoteNameE", "tensorrt_llm::executor::kv_cache::TransferRequest::mRemoteName"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest9mSrcDescsE", "tensorrt_llm::executor::kv_cache::TransferRequest::mSrcDescs"], [0, 5, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache15TransferRequest12mSyncMessageE", "tensorrt_llm::executor::kv_cache::TransferRequest::mSyncMessage"], [0, 2, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache14TransferStatusE", "tensorrt_llm::executor::kv_cache::TransferStatus"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache14TransferStatus11isCompletedEv", "tensorrt_llm::executor::kv_cache::TransferStatus::isCompleted"], [0, 3, 1, "_CPPv4NK12tensorrt_llm8executor8kv_cache14TransferStatus4waitEv", "tensorrt_llm::executor::kv_cache::TransferStatus::wait"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor8kv_cache14TransferStatusD0Ev", "tensorrt_llm::executor::kv_cache::TransferStatus::~TransferStatus"], [0, 3, 1, "_CPPv4IDpEN12tensorrt_llm8executor8kv_cache17makeTransferAgentENSt10unique_ptrI17BaseTransferAgentEERKNSt6stringEDpRR4Args", "tensorrt_llm::executor::kv_cache::makeTransferAgent"], [0, 8, 1, "_CPPv4IDpEN12tensorrt_llm8executor8kv_cache17makeTransferAgentENSt10unique_ptrI17BaseTransferAgentEERKNSt6stringEDpRR4Args", "tensorrt_llm::executor::kv_cache::makeTransferAgent::Args"], [0, 4, 1, "_CPPv4IDpEN12tensorrt_llm8executor8kv_cache17makeTransferAgentENSt10unique_ptrI17BaseTransferAgentEERKNSt6stringEDpRR4Args", "tensorrt_llm::executor::kv_cache::makeTransferAgent::args"], [0, 4, 1, "_CPPv4IDpEN12tensorrt_llm8executor8kv_cache17makeTransferAgentENSt10unique_ptrI17BaseTransferAgentEERKNSt6stringEDpRR4Args", "tensorrt_llm::executor::kv_cache::makeTransferAgent::backend"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<::os"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE21ContextChunkingPolicy", "tensorrt_llm::executor::operator<<::policy"], [0, 4, 1, "_CPPv4N12tensorrt_llm8executorlsERNSt7ostreamE23CapacitySchedulerPolicy", "tensorrt_llm::executor::operator<<::policy"], [0, 3, 1, "_CPPv4N12tensorrt_llm8executor7versionEv", "tensorrt_llm::executor::version"], [1, 1, 1, "_CPPv4N12tensorrt_llm6layersE", "tensorrt_llm::layers"], [0, 1, 1, "_CPPv4N12tensorrt_llm3mpiE", "tensorrt_llm::mpi"], [0, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [0, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtimeE", "tensorrt_llm::runtime"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffersE", "tensorrt_llm::runtime::AllReduceBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::fakeBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers16AllReduceBuffersE10SizeType3210SizeType3210SizeType3210SizeType32RK13BufferManagerRK11WorldConfigKb", "tensorrt_llm::runtime::AllReduceBuffers::AllReduceBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9TensorPtrE", "tensorrt_llm::runtime::AllReduceBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers18mAllReduceCommPtrsE", "tensorrt_llm::runtime::AllReduceBuffers::mAllReduceCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers9mFlagPtrsE", "tensorrt_llm::runtime::AllReduceBuffers::mFlagPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16AllReduceBuffers17mIpcMemoryHandlesE", "tensorrt_llm::runtime::AllReduceBuffers::mIpcMemoryHandles"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataTypeE", "tensorrt_llm::runtime::BufferDataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::_unsigned"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::dataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType14BufferDataTypeEN8nvinfer18DataTypeEbb", "tensorrt_llm::runtime::BufferDataType::BufferDataType::pointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType11getDataTypeEv", "tensorrt_llm::runtime::BufferDataType::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType7getSizeEv", "tensorrt_llm::runtime::BufferDataType::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType13getSizeInBitsEv", "tensorrt_llm::runtime::BufferDataType::getSizeInBits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType9isPointerEv", "tensorrt_llm::runtime::BufferDataType::isPointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataType10isUnsignedEv", "tensorrt_llm::runtime::BufferDataType::isUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType15kTrtPointerTypeE", "tensorrt_llm::runtime::BufferDataType::kTrtPointerType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mDataTypeE", "tensorrt_llm::runtime::BufferDataType::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType8mPointerE", "tensorrt_llm::runtime::BufferDataType::mPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14BufferDataType9mUnsignedE", "tensorrt_llm::runtime::BufferDataType::mUnsigned"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14BufferDataTypecvN8nvinfer18DataTypeEEv", "tensorrt_llm::runtime::BufferDataType::operator nvinfer1::DataType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerE", "tensorrt_llm::runtime::BufferManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13BufferManagerE13CudaStreamPtrb", "tensorrt_llm::runtime::BufferManager::BufferManager::trimPool"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager14CudaMemPoolPtrE", "tensorrt_llm::runtime::BufferManager::CudaMemPoolPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager13CudaStreamPtrE", "tensorrt_llm::runtime::BufferManager::CudaStreamPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10IBufferPtrE", "tensorrt_llm::runtime::BufferManager::IBufferPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10ITensorPtrE", "tensorrt_llm::runtime::BufferManager::ITensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::dims"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::size"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::type"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8allocateE10MemoryTypeNSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::allocate::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::dst"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::dstType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferPv10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyERK7IBufferR7IBuffer", "tensorrt_llm::runtime::BufferManager::copy::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager4copyEPKvR7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copy::srcType"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::memoryType"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10IBufferPtrRKNSt6vectorI1TEE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrP1TN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime13BufferManager8copyFromE10ITensorPtrRKNSt6vectorI1TEEN8nvinfer14DimsE10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7IBuffer10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager8copyFromERK7ITensor10MemoryType", "tensorrt_llm::runtime::BufferManager::copyFrom::src"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager3cpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::cpu::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyBufferE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyBuffer::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor::memoryType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager11emptyTensorE10MemoryTypeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::emptyTensor::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager9getStreamEv", "tensorrt_llm::runtime::BufferManager::getStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::dims"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::size"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::type"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager3gpuENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpu::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7gpuSyncENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::gpuSync::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7ipcNvlsENSt3setIiEEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::ipcNvls::type"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10kBYTE_TYPEE", "tensorrt_llm::runtime::BufferManager::kBYTE_TYPE"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager5mPoolE", "tensorrt_llm::runtime::BufferManager::mPool"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7mStreamE", "tensorrt_llm::runtime::BufferManager::mStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager9mTrimPoolE", "tensorrt_llm::runtime::BufferManager::mTrimPool"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager7managedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::managed::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolFreeEv", "tensorrt_llm::runtime::BufferManager::memoryPoolFree"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager18memoryPoolReservedEv", "tensorrt_llm::runtime::BufferManager::memoryPoolReserved"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", "tensorrt_llm::runtime::BufferManager::memoryPoolTrimTo"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager16memoryPoolTrimToENSt6size_tE", "tensorrt_llm::runtime::BufferManager::memoryPoolTrimTo::size"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager14memoryPoolUsedEv", "tensorrt_llm::runtime::BufferManager::memoryPoolUsed"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager6pinnedENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinned::type"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolEN8nvinfer14DimsEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManager10pinnedPoolENSt6size_tEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::BufferManager::pinnedPool::type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem::buffer"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager6setMemER7IBuffer7int32_t", "tensorrt_llm::runtime::BufferManager::setMem::value"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", "tensorrt_llm::runtime::BufferManager::setZero"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13BufferManager7setZeroER7IBuffer", "tensorrt_llm::runtime::BufferManager::setZero::buffer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13BufferManagerD0Ev", "tensorrt_llm::runtime::BufferManager::~BufferManager"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", "tensorrt_llm::runtime::BufferRange"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange4BaseE", "tensorrt_llm::runtime::BufferRange::Base"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::U"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::U"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI1UEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeERK7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tIXntNSt10is_const_vI1UEEEbEEEN12tensorrt_llm7runtime11BufferRange11BufferRangeER7IBuffer", "tensorrt_llm::runtime::BufferRange::BufferRange::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11BufferRange11BufferRangeEP1T9size_type", "tensorrt_llm::runtime::BufferRange::BufferRange::size"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime11BufferRangeE", "tensorrt_llm::runtime::BufferRange::T"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEventE", "tensorrt_llm::runtime::CudaEvent"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", "tensorrt_llm::runtime::CudaEvent::CudaEvent"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent::event"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventEj", "tensorrt_llm::runtime::CudaEvent::CudaEvent::flags"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent9CudaEventE7pointerb", "tensorrt_llm::runtime::CudaEvent::CudaEvent::ownsEvent"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7DeleterE", "tensorrt_llm::runtime::CudaEvent::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEv", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaEvent::Deleter::Deleter::ownsEvent"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7Deleter10mOwnsEventE", "tensorrt_llm::runtime::CudaEvent::Deleter::mOwnsEvent"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", "tensorrt_llm::runtime::CudaEvent::Deleter::operator()"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent7DeleterclE7pointer", "tensorrt_llm::runtime::CudaEvent::Deleter::operator()::event"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent8EventPtrE", "tensorrt_llm::runtime::CudaEvent::EventPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent12element_typeE", "tensorrt_llm::runtime::CudaEvent::element_type"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent3getEv", "tensorrt_llm::runtime::CudaEvent::get"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent6mEventE", "tensorrt_llm::runtime::CudaEvent::mEvent"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaEvent::pointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9CudaEvent11synchronizeEv", "tensorrt_llm::runtime::CudaEvent::synchronize"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStreamE", "tensorrt_llm::runtime::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::device"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream::flags"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::ownsStream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamEji", "tensorrt_llm::runtime::CudaStream::CudaStream::priority"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::CudaStream::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream10CudaStreamE12cudaStream_tib", "tensorrt_llm::runtime::CudaStream::CudaStream::stream"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7DeleterE", "tensorrt_llm::runtime::CudaStream::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEv", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter7DeleterEb", "tensorrt_llm::runtime::CudaStream::Deleter::Deleter::ownsStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7Deleter11mOwnsStreamE", "tensorrt_llm::runtime::CudaStream::Deleter::mOwnsStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::Deleter::operator()"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream7DeleterclE12cudaStream_t", "tensorrt_llm::runtime::CudaStream::Deleter::operator()::stream"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream9StreamPtrE", "tensorrt_llm::runtime::CudaStream::StreamPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream3getEv", "tensorrt_llm::runtime::CudaStream::get"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream9getDeviceEv", "tensorrt_llm::runtime::CudaStream::getDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mDeviceE", "tensorrt_llm::runtime::CudaStream::mDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10CudaStream7mStreamE", "tensorrt_llm::runtime::CudaStream::mStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::record"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::record"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::record::event"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream6recordERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::record::event"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream11synchronizeEv", "tensorrt_llm::runtime::CudaStream::synchronize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::wait"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::wait"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitEN9CudaEvent7pointerE", "tensorrt_llm::runtime::CudaStream::wait::event"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10CudaStream4waitERK9CudaEvent", "tensorrt_llm::runtime::CudaStream::wait::event"], [1, 2, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kDataType"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kIsPointer"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_b_bEN12tensorrt_llm7runtime14DataTypeTraitsE", "tensorrt_llm::runtime::DataTypeTraits::kIsUnsigned"], [1, 2, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::kDataType"], [1, 8, 1, "_CPPv4I_N8nvinfer18DataTypeE_bEN12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsI9kDataType9kUnsignedXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<kDataType, kUnsigned, true>::type"], [1, 2, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>"], [1, 8, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kBOOLE9kUnsignedE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kBOOL, kUnsigned>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kFLOATEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kFLOAT>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kHALFEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kHALF>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32, true>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT32EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT32>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EXL1EEE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64, true>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kINT64EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT64>::type"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType5kINT8EE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kINT8>::type"], [1, 2, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>"], [1, 8, 1, "_CPPv4I_bEN12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedEE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::kUnsigned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4nameE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::name"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4sizeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::size"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DataTypeTraitsIN8nvinfer18DataType6kUINT8E9kUnsignedE4typeE", "tensorrt_llm::runtime::DataTypeTraits<nvinfer1::DataType::kUINT8, kUnsigned>::type"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInputE", "tensorrt_llm::runtime::DecodingInput"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::endIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::maxLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13DecodingInputE10SizeType3210SizeType3210SizeType3210SizeType3214TensorConstPtr9TensorPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::DecodingInput::sinkTokenLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedPathIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::acceptedTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::chunkedContextNextTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::lastDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftLens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::nextDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs11EagleInputsE14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr14TensorConstPtr", "tensorrt_llm::runtime::DecodingInput::EagleInputs::EagleInputs::seqSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs12acceptedLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15acceptedPathIdsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedPathIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14acceptedTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::acceptedTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::chunkedContextNextTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13lastDraftLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14lastDraftPathsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15lastDraftTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::lastDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs13nextDraftLensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs14nextDraftPathsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11EagleInputs8seqSlotsE", "tensorrt_llm::runtime::DecodingInput::EagleInputs::seqSlots"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::bestPathIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15bestPathLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::bestPathLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16lastDraftIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15lastDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21lastGenerationLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs19lastPositionIdsBaseE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::lastPositionIdsBase"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs5masksE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::masks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs18maxGenLengthDeviceE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::maxGenLengthDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs16nextDraftIndicesE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextDraftProbsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs14nextFlatTokensE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextFlatTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs21nextGenerationLengthsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::nextGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs17packedPositionIdsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::packedPositionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExplicitDraftTokensInputs8seqSlotsE", "tensorrt_llm::runtime::DecodingInput::ExplicitDraftTokensInputs::seqSlots"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs17constantThresholdE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::constantThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11draftLogitsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs10draftProbsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs13draftTokenIdsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::draftTokenIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14numDraftTokensE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::numDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18numDraftTokensHostE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::numDraftTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs4stepE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::step"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs11targetProbsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::targetProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs14useDraftLogitsE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useDraftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs18useDraftLogitsHostE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useDraftLogitsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25ExternalDraftTokensInputs28useRandomAcceptanceThresholdE", "tensorrt_llm::runtime::DecodingInput::ExternalDraftTokensInputs::useRandomAcceptanceThreshold"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputsE", "tensorrt_llm::runtime::DecodingInput::LookaheadInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15LookaheadInputs13tokensPerStepE", "tensorrt_llm::runtime::DecodingInput::LookaheadInputs::tokensPerStep"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs22medusaCurTokensPerStepE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaCurTokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs12medusaLogitsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs11medusaPathsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs25medusaTargetTokensPerStepE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaTargetTokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12MedusaInputs13medusaTreeIdsE", "tensorrt_llm::runtime::DecodingInput::MedusaInputs::medusaTreeIds"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14TensorConstPtrE", "tensorrt_llm::runtime::DecodingInput::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9TensorPtrE", "tensorrt_llm::runtime::DecodingInput::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsLensE", "tensorrt_llm::runtime::DecodingInput::badWordsLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13badWordsListsE", "tensorrt_llm::runtime::DecodingInput::badWordsLists"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12badWordsPtrsE", "tensorrt_llm::runtime::DecodingInput::badWordsPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9batchSizeE", "tensorrt_llm::runtime::DecodingInput::batchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10batchSlotsE", "tensorrt_llm::runtime::DecodingInput::batchSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput10beamWidthsE", "tensorrt_llm::runtime::DecodingInput::beamWidths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput16cacheIndirectionE", "tensorrt_llm::runtime::DecodingInput::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput11eagleInputsE", "tensorrt_llm::runtime::DecodingInput::eagleInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13embeddingBiasE", "tensorrt_llm::runtime::DecodingInput::embeddingBias"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6endIdsE", "tensorrt_llm::runtime::DecodingInput::endIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25explicitDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::explicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput25externalDraftTokensInputsE", "tensorrt_llm::runtime::DecodingInput::externalDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13finishReasonsE", "tensorrt_llm::runtime::DecodingInput::finishReasons"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15generationStepsE", "tensorrt_llm::runtime::DecodingInput::generationSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput7lengthsE", "tensorrt_llm::runtime::DecodingInput::lengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput6logitsE", "tensorrt_llm::runtime::DecodingInput::logits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9logitsVecE", "tensorrt_llm::runtime::DecodingInput::logitsVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15lookaheadInputsE", "tensorrt_llm::runtime::DecodingInput::lookaheadInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput18maxAttentionWindowE", "tensorrt_llm::runtime::DecodingInput::maxAttentionWindow"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14maxBadWordsLenE", "tensorrt_llm::runtime::DecodingInput::maxBadWordsLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput9maxLengthE", "tensorrt_llm::runtime::DecodingInput::maxLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15maxStopWordsLenE", "tensorrt_llm::runtime::DecodingInput::maxStopWordsLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput12medusaInputsE", "tensorrt_llm::runtime::DecodingInput::medusaInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput17noRepeatNgramSizeE", "tensorrt_llm::runtime::DecodingInput::noRepeatNgramSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput19sequenceLimitLengthE", "tensorrt_llm::runtime::DecodingInput::sequenceLimitLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput15sinkTokenLengthE", "tensorrt_llm::runtime::DecodingInput::sinkTokenLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput4stepE", "tensorrt_llm::runtime::DecodingInput::step"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsLensE", "tensorrt_llm::runtime::DecodingInput::stopWordsLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput14stopWordsListsE", "tensorrt_llm::runtime::DecodingInput::stopWordsLists"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13DecodingInput13stopWordsPtrsE", "tensorrt_llm::runtime::DecodingInput::stopWordsPtrs"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutputE", "tensorrt_llm::runtime::DecodingOutput"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypothesesE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses10batchDonesE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::batchDones"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses14cumLogProbsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::cumLogProbsCBA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::empty"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5emptyERK13BufferManager", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::empty::manager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init::endId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses4initERK13BufferManager11TokenIdType", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::init::manager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11logProbsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::logProbsCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18minNormedScoresCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::minNormedScoresCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses15normedScoresCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::normedScoresCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses11numBeamsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::numBeamsCBA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses12outputIdsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::outputIdsCBA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7releaseEv", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::release"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::beamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::reshape::maxSequenceLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses18sequenceLengthsCBAE", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::sequenceLengthsCBA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice::batchIndex"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14DecodingOutput14BeamHypotheses5sliceE10SizeType3210SizeType32", "tensorrt_llm::runtime::DecodingOutput::BeamHypotheses::slice::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput::gatheredIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14DecodingOutputE9TensorPtr9TensorPtr", "tensorrt_llm::runtime::DecodingOutput::DecodingOutput::ids"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputsE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs21acceptedLengthsCumSumE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::acceptedLengthsCumSum"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs17acceptedTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::acceptedTokensLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs15nextDraftTokensE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18nextDraftTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::nextDraftTokensLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs12pathsOffsetsE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::pathsOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26SpeculativeDecodingOutputs18prevDraftTokensLenE", "tensorrt_llm::runtime::DecodingOutput::SpeculativeDecodingOutputs::prevDraftTokensLen"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9TensorPtrE", "tensorrt_llm::runtime::DecodingOutput::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14beamHypothesesE", "tensorrt_llm::runtime::DecodingOutput::beamHypotheses"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16cacheIndirectionE", "tensorrt_llm::runtime::DecodingOutput::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11cumLogProbsE", "tensorrt_llm::runtime::DecodingOutput::cumLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12eagleBuffersE", "tensorrt_llm::runtime::DecodingOutput::eagleBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26explicitDraftTokensBuffersE", "tensorrt_llm::runtime::DecodingOutput::explicitDraftTokensBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13finishReasonsE", "tensorrt_llm::runtime::DecodingOutput::finishReasons"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11finishedSumE", "tensorrt_llm::runtime::DecodingOutput::finishedSum"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput11gatheredIdsE", "tensorrt_llm::runtime::DecodingOutput::gatheredIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput3idsE", "tensorrt_llm::runtime::DecodingOutput::ids"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput17kNegativeInfinityE", "tensorrt_llm::runtime::DecodingOutput::kNegativeInfinity"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput7lengthsE", "tensorrt_llm::runtime::DecodingOutput::lengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput8logProbsE", "tensorrt_llm::runtime::DecodingOutput::logProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput13logProbsTiledE", "tensorrt_llm::runtime::DecodingOutput::logProbsTiled"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput16lookaheadOutputsE", "tensorrt_llm::runtime::DecodingOutput::lookaheadOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9newTokensE", "tensorrt_llm::runtime::DecodingOutput::newTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput14newTokensStepsE", "tensorrt_llm::runtime::DecodingOutput::newTokensSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput12newTokensVecE", "tensorrt_llm::runtime::DecodingOutput::newTokensVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput9parentIdsE", "tensorrt_llm::runtime::DecodingOutput::parentIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14DecodingOutput26speculativeDecodingOutputsE", "tensorrt_llm::runtime::DecodingOutput::speculativeDecodingOutputs"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", "tensorrt_llm::runtime::DeviceAllocationNvls"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls20DeviceAllocationNvlsEv", "tensorrt_llm::runtime::DeviceAllocationNvls::DeviceAllocationNvls"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime20DeviceAllocationNvlsE", "tensorrt_llm::runtime::DeviceAllocationNvls::T"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls9_capacityE", "tensorrt_llm::runtime::DeviceAllocationNvls::_capacity"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls7_handleE", "tensorrt_llm::runtime::DeviceAllocationNvls::_handle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls4freeEv", "tensorrt_llm::runtime::DeviceAllocationNvls::free"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls11getCapacityEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getCapacity"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls21getIpcUnicastPointersEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getIpcUnicastPointers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls19getMulticastPointerEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getMulticastPointer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20DeviceAllocationNvls17getUnicastPointerEv", "tensorrt_llm::runtime::DeviceAllocationNvls::getUnicastPointer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvls5resetE6size_tNSt3setIiEE", "tensorrt_llm::runtime::DeviceAllocationNvls::reset::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20DeviceAllocationNvlsD0Ev", "tensorrt_llm::runtime::DeviceAllocationNvls::~DeviceAllocationNvls"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffersE", "tensorrt_llm::runtime::EagleBuffers"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9BufferPtrE", "tensorrt_llm::runtime::EagleBuffers::BufferPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::decodingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12EagleBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN8executor14DecodingConfigE", "tensorrt_llm::runtime::EagleBuffers::EagleBuffers::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs12acceptedLensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13acceptedPathsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14acceptedTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::acceptedTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::chunkedContextNextTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs13nextDraftLensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs14nextDraftPathsE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13EngineOutputs15nextDraftTokensE", "tensorrt_llm::runtime::EagleBuffers::EngineOutputs::nextDraftTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7ITensorE", "tensorrt_llm::runtime::EagleBuffers::ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6InputsE", "tensorrt_llm::runtime::EagleBuffers::Inputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22allLayersDraftTokenIdsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersDraftTokenIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33allLayersDraftTokenIdsPredecessorE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersDraftTokenIdsPredecessor"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15allLayersScoresE", "tensorrt_llm::runtime::EagleBuffers::Inputs::allLayersScores"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs24chunkedContextNextTokensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::chunkedContextNextTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs6createE10SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::EagleBuffers::Inputs::create::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20currentExpandIndicesE", "tensorrt_llm::runtime::EagleBuffers::Inputs::currentExpandIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs9draftLensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftLens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10draftPathsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14draftPathsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftPathsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs11draftTokensE", "tensorrt_llm::runtime::EagleBuffers::Inputs::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs22dynamicTreeMaxTopKHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::dynamicTreeMaxTopKHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetCtxContextLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxContextLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetCtxPastKeyValueLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxPastKeyValueLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetCtxRequestTypesHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetCtxRequestTypesHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29eagleNetGenContextLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenContextLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs34eagleNetGenPastKeyValueLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenPastKeyValueLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27eagleNetGenRequestTypesHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::eagleNetGenRequestTypesHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18inputGenTokensHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::inputGenTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs14posteriorAlphaE", "tensorrt_llm::runtime::EagleBuffers::Inputs::posteriorAlpha"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18posteriorThresholdE", "tensorrt_llm::runtime::EagleBuffers::Inputs::posteriorThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs10prevScoresE", "tensorrt_llm::runtime::EagleBuffers::Inputs::prevScores"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs16randomDataSampleE", "tensorrt_llm::runtime::EagleBuffers::Inputs::randomDataSample"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs20randomDataValidationE", "tensorrt_llm::runtime::EagleBuffers::Inputs::randomDataValidation"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs29specDecodingGenerationLengthsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs33specDecodingGenerationLengthsHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingGenerationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs23specDecodingPackedMasksE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs27specDecodingPositionOffsetsE", "tensorrt_llm::runtime::EagleBuffers::Inputs::specDecodingPositionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs12temperaturesE", "tensorrt_llm::runtime::EagleBuffers::Inputs::temperatures"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs18useDynamicTreeHostE", "tensorrt_llm::runtime::EagleBuffers::Inputs::useDynamicTreeHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers6Inputs15useSpecDecodingE", "tensorrt_llm::runtime::EagleBuffers::Inputs::useSpecDecoding"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13LlmRequestPtrE", "tensorrt_llm::runtime::EagleBuffers::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13RequestVectorE", "tensorrt_llm::runtime::EagleBuffers::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers10SizeType32E", "tensorrt_llm::runtime::EagleBuffers::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorMapE", "tensorrt_llm::runtime::EagleBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers9TensorPtrE", "tensorrt_llm::runtime::EagleBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers28chunkedContextNextTokensHostE", "tensorrt_llm::runtime::EagleBuffers::chunkedContextNextTokensHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers23cumSumGenerationLengthsE", "tensorrt_llm::runtime::EagleBuffers::cumSumGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers12engineInputsE", "tensorrt_llm::runtime::EagleBuffers::engineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers13engineOutputsE", "tensorrt_llm::runtime::EagleBuffers::engineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18greedySamplingHostE", "tensorrt_llm::runtime::EagleBuffers::greedySamplingHost"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::insertInputTensors::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26mDefaultPosteriorThresholdE", "tensorrt_llm::runtime::EagleBuffers::mDefaultPosteriorThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers17mDoGreedySamplingE", "tensorrt_llm::runtime::EagleBuffers::mDoGreedySampling"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers19maxGenerationLengthE", "tensorrt_llm::runtime::EagleBuffers::maxGenerationLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers18posteriorAlphaHostE", "tensorrt_llm::runtime::EagleBuffers::posteriorAlphaHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers22posteriorThresholdHostE", "tensorrt_llm::runtime::EagleBuffers::posteriorThresholdHost"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::EagleBuffers::reshape::numGenSequences"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers21scanReduceTempStorageE", "tensorrt_llm::runtime::EagleBuffers::scanReduceTempStorage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12EagleBuffers26scanReduceTempStorageBytesE", "tensorrt_llm::runtime::EagleBuffers::scanReduceTempStorageBytes"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::contextRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::contextRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::decoderBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::draftBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::eagleModule"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::genRequests"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::genRequests"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime12EagleBuffers13setFromInputsEvRK13RequestVectorRK13RequestVector10SizeType32RK7ITensorRKN12EagleBuffers6InputsERKN7runtime11EagleModuleERKN7runtime13BufferManagerE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::vocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime12EagleBuffers13setFromInputsERK13RequestVectorRK13RequestVectorRKN7runtime7ITensorERK7ITensorRKN12EagleBuffers6InputsERKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::EagleBuffers::setFromInputs::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModuleE", "tensorrt_llm::runtime::EagleModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleEv", "tensorrt_llm::runtime::EagleModule::EagleModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::maxNonLeafNodesPerLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule11EagleModuleE10SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::EagleModule::EagleModule::numTransformersLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule22getDefaultEagleChoicesEv", "tensorrt_llm::runtime::EagleModule::getDefaultEagleChoices"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule26getMaxNonLeafNodesPerLayerEv", "tensorrt_llm::runtime::EagleModule::getMaxNonLeafNodesPerLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11EagleModule23getNumTransformerLayersEv", "tensorrt_llm::runtime::EagleModule::getNumTransformerLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule20mDefaultEagleChoicesE", "tensorrt_llm::runtime::EagleModule::mDefaultEagleChoices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule24mMaxNonLeafNodesPerLayerE", "tensorrt_llm::runtime::EagleModule::mMaxNonLeafNodesPerLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11EagleModule21mNumTransformersLayerE", "tensorrt_llm::runtime::EagleModule::mNumTransformersLayer"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffersE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9BufferPtrE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::BufferPtr"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs15positionOffsetsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs::positionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12EngineInputs18requestTypesDeviceE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineInputs::requestTypesDevice"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::bestPathIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15bestPathLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::bestPathLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs5masksE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::masks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs11maxGenTokenE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::maxGenToken"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs16nextDraftIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextDraftProbsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs15nextDraftTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs14nextFlatTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextFlatTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs21nextGenerationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs19nextPositionOffsetsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::nextPositionOffsets"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs17packedPositionIdsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::packedPositionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13EngineOutputs13totalGenTokenE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::EngineOutputs::totalGenToken"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers26ExplicitDraftTokensBuffersE10SizeType3210SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ExplicitDraftTokensBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7ITensorE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs6createE10SizeType32RKN7runtime13BufferManagerERKN7runtime11ModelConfigERKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::create::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12draftIndicesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftIndices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs10draftProbsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11draftTokensE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs17generationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::generationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs21generationLengthsHostE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::generationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16maxGenLengthHostE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::maxGenLengthHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11packedMasksE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::packedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs11positionIdsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::positionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15positionIdsBaseE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::positionIdsBase"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs16randomDataSampleE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::randomDataSample"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs20randomDataValidationE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::randomDataValidation"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs12temperaturesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::temperatures"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers6Inputs15useSpecDecodingE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::Inputs::useSpecDecoding"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers10SizeType32E", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorMapE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers9TensorPtrE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers23cumSumGenerationLengthsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::cumSumGenerationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers12engineInputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::engineInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13engineOutputsE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::engineOutputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers18insertInputTensorsER9TensorMapR9TensorMapRKN7runtime11WorldConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::insertInputTensors::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers7reshapeE10SizeType3210SizeType32RKN7runtime11ModelConfigE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::reshape::numGenSequences"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers15scanTempStorageE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::scanTempStorage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26ExplicitDraftTokensBuffers20scanTempStorageBytesE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::scanTempStorageBytes"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::T"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::contextPositionIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::contextPositionIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::decoderBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::draftBuffers"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::explicitDraftTokensModule"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::manager"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::stream"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::stream"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsEv10SizeType3210SizeType3210SizeType32RK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime25ExplicitDraftTokensModuleERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::vocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime26ExplicitDraftTokensBuffers13setFromInputsE10SizeType3210SizeType32RKN7runtime7ITensorERK7ITensorRKN26ExplicitDraftTokensBuffers6InputsERK7ITensorRKN7runtime11ModelConfigERKN7runtime11WorldConfigERKN7runtime13BufferManagerERKN7runtime10CudaStreamE", "tensorrt_llm::runtime::ExplicitDraftTokensBuffers::setFromInputs::worldConfig"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", "tensorrt_llm::runtime::GenericPromptTuningParams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::embeddingTable"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::tasks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams25GenericPromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::GenericPromptTuningParams::GenericPromptTuningParams::vocabSize"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams10SizeType32E", "tensorrt_llm::runtime::GenericPromptTuningParams::SizeType32"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime25GenericPromptTuningParamsE", "tensorrt_llm::runtime::GenericPromptTuningParams::TTensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9TensorPtrE", "tensorrt_llm::runtime::GenericPromptTuningParams::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams14embeddingTableE", "tensorrt_llm::runtime::GenericPromptTuningParams::embeddingTable"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams19promptTuningEnabledE", "tensorrt_llm::runtime::GenericPromptTuningParams::promptTuningEnabled"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams5tasksE", "tensorrt_llm::runtime::GenericPromptTuningParams::tasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25GenericPromptTuningParams9vocabSizeE", "tensorrt_llm::runtime::GenericPromptTuningParams::vocabSize"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", "tensorrt_llm::runtime::GptDecoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13CudaStreamPtrE", "tensorrt_llm::runtime::GptDecoder::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::speculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::vocabSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10GptDecoderERKN8executor12DecodingModeE6size_t6size_t6size_t6size_t6size_tRK13CudaStreamPtrNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::GptDecoder::GptDecoder::vocabSizePadded"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10GptDecoderE", "tensorrt_llm::runtime::GptDecoder::T"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder9TensorPtrE", "tensorrt_llm::runtime::GptDecoder::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::GptDecoder::disableLookahead::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::GptDecoder::forwardSync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder17getSamplingConfigEv", "tensorrt_llm::runtime::GptDecoder::getSamplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder23mDecodingLayerWorkspaceE", "tensorrt_llm::runtime::GptDecoder::mDecodingLayerWorkspace"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mDecodingModeE", "tensorrt_llm::runtime::GptDecoder::mDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder19mDynamicDecodeLayerE", "tensorrt_llm::runtime::GptDecoder::mDynamicDecodeLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder8mManagerE", "tensorrt_llm::runtime::GptDecoder::mManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder13mMaxBatchSizeE", "tensorrt_llm::runtime::GptDecoder::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder15mSamplingConfigE", "tensorrt_llm::runtime::GptDecoder::mSamplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder10mVocabSizeE", "tensorrt_llm::runtime::GptDecoder::mVocabSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder16mVocabSizePaddedE", "tensorrt_llm::runtime::GptDecoder::mVocabSizePadded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::requests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10GptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::GptDecoder::setup::samplingConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatchedE", "tensorrt_llm::runtime::GptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13CudaStreamPtrE", "tensorrt_llm::runtime::GptDecoderBatched::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::speculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched17GptDecoderBatchedE13CudaStreamPtrRK23SpeculativeDecodingModeN8nvinfer18DataTypeE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderBatched::stream"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13GptDecoderPtrE", "tensorrt_llm::runtime::GptDecoderBatched::GptDecoderPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13LlmRequestPtrE", "tensorrt_llm::runtime::GptDecoderBatched::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13RequestVectorE", "tensorrt_llm::runtime::GptDecoderBatched::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14SharedConstPtrE", "tensorrt_llm::runtime::GptDecoderBatched::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched9TensorPtrE", "tensorrt_llm::runtime::GptDecoderBatched::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::GptDecoderBatched::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::batchSlot"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::decoderState"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::samplingConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::GptDecoderBatched::finalize::streaming"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forward::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15forwardDispatchERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::forwardDispatch::output"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getBufferManagerEv", "tensorrt_llm::runtime::GptDecoderBatched::getBufferManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched15getDecoderStateEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched16getDecoderStreamEv", "tensorrt_llm::runtime::GptDecoderBatched::getDecoderStream"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime17GptDecoderBatched20getUnderlyingDecoderEv", "tensorrt_llm::runtime::GptDecoderBatched::getUnderlyingDecoder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mBufferManagerE", "tensorrt_llm::runtime::GptDecoderBatched::mBufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched8mDecoderE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched13mDecoderStateE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoderState"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mDecoderStreamE", "tensorrt_llm::runtime::GptDecoderBatched::mDecoderStream"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14mRuntimeStreamE", "tensorrt_llm::runtime::GptDecoderBatched::mRuntimeStream"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14prepareForwardE10SizeType32RN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::prepareForward::step"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setEagleInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched14setEagleInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setEagleInputs::input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setExplicitDraftTokensInputs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched28setExplicitDraftTokensInputsERKN13decoder_batch5InputE", "tensorrt_llm::runtime::GptDecoderBatched::setExplicitDraftTokensInputs::input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::maxTokensPerStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17GptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::GptDecoderBatched::setup::worldConfig"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfigE", "tensorrt_llm::runtime::GptJsonConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::name"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::precision"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::runtimeDefaults"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::tensorParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig13GptJsonConfigENSt6stringENSt6stringENSt6stringE10SizeType3210SizeType3210SizeType3210SizeType3211ModelConfigNSt8optionalI15RuntimeDefaultsEE", "tensorrt_llm::runtime::GptJsonConfig::GptJsonConfig::version"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", "tensorrt_llm::runtime::GptJsonConfig::engineFilename"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::model"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfig", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::worldConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14engineFilenameERK11WorldConfigRKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::engineFilename::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig21getContextParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getContextParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getGpusPerNodeEv", "tensorrt_llm::runtime::GptJsonConfig::getGpusPerNode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig14getModelConfigEv", "tensorrt_llm::runtime::GptJsonConfig::getModelConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig21getModelConfigMutableEv", "tensorrt_llm::runtime::GptJsonConfig::getModelConfigMutable"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig7getNameEv", "tensorrt_llm::runtime::GptJsonConfig::getName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig22getPipelineParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getPipelineParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getPrecisionEv", "tensorrt_llm::runtime::GptJsonConfig::getPrecision"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig18getRuntimeDefaultsEv", "tensorrt_llm::runtime::GptJsonConfig::getRuntimeDefaults"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig20getTensorParallelismEv", "tensorrt_llm::runtime::GptJsonConfig::getTensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig10getVersionEv", "tensorrt_llm::runtime::GptJsonConfig::getVersion"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime13GptJsonConfig12getWorldSizeEv", "tensorrt_llm::runtime::GptJsonConfig::getWorldSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig19mContextParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mContextParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mGpusPerNodeE", "tensorrt_llm::runtime::GptJsonConfig::mGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig12mModelConfigE", "tensorrt_llm::runtime::GptJsonConfig::mModelConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5mNameE", "tensorrt_llm::runtime::GptJsonConfig::mName"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig20mPipelineParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mPipelineParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig10mPrecisionE", "tensorrt_llm::runtime::GptJsonConfig::mPrecision"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig16mRuntimeDefaultsE", "tensorrt_llm::runtime::GptJsonConfig::mRuntimeDefaults"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig18mTensorParallelismE", "tensorrt_llm::runtime::GptJsonConfig::mTensorParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig8mVersionE", "tensorrt_llm::runtime::GptJsonConfig::mVersion"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", "tensorrt_llm::runtime::GptJsonConfig::parse"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt6stringE", "tensorrt_llm::runtime::GptJsonConfig::parse::json"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERNSt7istreamE", "tensorrt_llm::runtime::GptJsonConfig::parse::json"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13GptJsonConfig5parseERKNSt10filesystem4pathE", "tensorrt_llm::runtime::GptJsonConfig::parse::path"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferE", "tensorrt_llm::runtime::IBuffer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer8DataTypeE", "tensorrt_llm::runtime::IBuffer::DataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferERK7IBuffer", "tensorrt_llm::runtime::IBuffer::IBuffer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7IBufferEv", "tensorrt_llm::runtime::IBuffer::IBuffer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::IBuffer::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9SharedPtrE", "tensorrt_llm::runtime::IBuffer::SharedPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer14UniqueConstPtrE", "tensorrt_llm::runtime::IBuffer::UniqueConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer9UniquePtrE", "tensorrt_llm::runtime::IBuffer::UniquePtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataEv", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataEv", "tensorrt_llm::runtime::IBuffer::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data::index"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer4dataENSt6size_tE", "tensorrt_llm::runtime::IBuffer::data::index"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getCapacityEv", "tensorrt_llm::runtime::IBuffer::getCapacity"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer11getDataTypeEv", "tensorrt_llm::runtime::IBuffer::getDataType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", "tensorrt_llm::runtime::IBuffer::getDataTypeName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer15getDataTypeNameEv", "tensorrt_llm::runtime::IBuffer::getDataTypeName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer15getDataTypeNameE8DataType", "tensorrt_llm::runtime::IBuffer::getDataTypeName::dataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer13getMemoryTypeEv", "tensorrt_llm::runtime::IBuffer::getMemoryType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer17getMemoryTypeNameEv", "tensorrt_llm::runtime::IBuffer::getMemoryTypeName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7getSizeEv", "tensorrt_llm::runtime::IBuffer::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer14getSizeInBytesEv", "tensorrt_llm::runtime::IBuffer::getSizeInBytes"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", "tensorrt_llm::runtime::IBuffer::memoryType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer10memoryTypeEPKv", "tensorrt_llm::runtime::IBuffer::memoryType::data"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferaSERK7IBuffer", "tensorrt_llm::runtime::IBuffer::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer7releaseEv", "tensorrt_llm::runtime::IBuffer::release"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", "tensorrt_llm::runtime::IBuffer::resize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer6resizeENSt6size_tE", "tensorrt_llm::runtime::IBuffer::resize::newSize"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::TConstPtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::slice::tensor"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", "tensorrt_llm::runtime::IBuffer::toBytes"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7IBuffer7toBytesENSt6size_tE", "tensorrt_llm::runtime::IBuffer::toBytes::size"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", "tensorrt_llm::runtime::IBuffer::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7IBuffer4viewE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtr", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4viewE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::IBuffer::view::tensor"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::capacity"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::capacity"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrP1TNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7IBuffer4wrapEPv8DataTypeNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::IBuffer::wrap::type"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7IBuffer4wrapE9UniquePtrRNSt6vectorI1TEE", "tensorrt_llm::runtime::IBuffer::wrap::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7IBufferD0Ev", "tensorrt_llm::runtime::IBuffer::~IBuffer"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderE", "tensorrt_llm::runtime::IGptDecoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder14TensorConstPtrE", "tensorrt_llm::runtime::IGptDecoder::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder9TensorPtrE", "tensorrt_llm::runtime::IGptDecoder::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::speculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::stream"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::vocabSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder6createERKN8executor12DecodingModeEN8nvinfer18DataTypeE6size_t6size_t6size_t6size_t6size_tRKN13BufferManager13CudaStreamPtrERKNSt10shared_ptrIK25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::IGptDecoder::create::vocabSizePadded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder16disableLookaheadERKNSt8optionalI14SamplingConfigEE10SizeType3214TensorConstPtr", "tensorrt_llm::runtime::IGptDecoder::disableLookahead::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder12forwardAsyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder11forwardSyncER14DecodingOutputRK13DecodingInput", "tensorrt_llm::runtime::IGptDecoder::forwardSync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder17getSamplingConfigEv", "tensorrt_llm::runtime::IGptDecoder::getSamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::requests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoder5setupERK14SamplingConfig6size_tRK14TensorConstPtrRKNSt8optionalI14DecodingOutputEERKNSt8optionalIKNSt6vectorIN13decoder_batch7RequestEEEEE", "tensorrt_llm::runtime::IGptDecoder::setup::samplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11IGptDecoderD0Ev", "tensorrt_llm::runtime::IGptDecoder::~IGptDecoder"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedE", "tensorrt_llm::runtime::IGptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13CudaStreamPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::CudaStreamPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched18IGptDecoderBatchedEv", "tensorrt_llm::runtime::IGptDecoderBatched::IGptDecoderBatched"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13LlmRequestPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched13RequestVectorE", "tensorrt_llm::runtime::IGptDecoderBatched::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched9TensorPtrE", "tensorrt_llm::runtime::IGptDecoderBatched::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead::batchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched16disableLookaheadERK13RequestVectorRK9TensorPtr", "tensorrt_llm::runtime::IGptDecoderBatched::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::batchSlot"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::decoderState"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::samplingConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime18IGptDecoderBatched8finalizeERKN7decoder12DecoderStateE10SizeType32RK14SamplingConfigb", "tensorrt_llm::runtime::IGptDecoderBatched::finalize::streaming"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched7forwardERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forward::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched12forwardAsyncERN13decoder_batch6OutputERKN13decoder_batch5InputE", "tensorrt_llm::runtime::IGptDecoderBatched::forwardAsync::output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::maxTokensPerStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::mode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatched5setupERKN8executor12DecodingModeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeERK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::IGptDecoderBatched::setup::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18IGptDecoderBatchedD0Ev", "tensorrt_llm::runtime::IGptDecoderBatched::~IGptDecoderBatched"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7ITensorE", "tensorrt_llm::runtime::ITensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9DimType64E", "tensorrt_llm::runtime::ITensor::DimType64"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorERK7ITensor", "tensorrt_llm::runtime::ITensor::ITensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7ITensorEv", "tensorrt_llm::runtime::ITensor::ITensor"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5ShapeE", "tensorrt_llm::runtime::ITensor::Shape"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::ITensor::SharedConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9SharedPtrE", "tensorrt_llm::runtime::ITensor::SharedPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9TensorMapE", "tensorrt_llm::runtime::ITensor::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor14UniqueConstPtrE", "tensorrt_llm::runtime::ITensor::UniqueConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9UniquePtrE", "tensorrt_llm::runtime::ITensor::UniquePtr"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor2atEN7ITensor14UniqueConstPtrERR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor2atE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::at::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", "tensorrt_llm::runtime::ITensor::castSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8castSizeE6size_t", "tensorrt_llm::runtime::ITensor::castSize::newSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN::sliceN"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8flattenNE9SharedPtrNSt7int64_tE", "tensorrt_llm::runtime::ITensor::flattenN::tensor"], [1, 3, 1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", "tensorrt_llm::runtime::ITensor::getDimension"], [1, 8, 1, "_CPPv4I_10SizeType32ENK12tensorrt_llm7runtime7ITensor12getDimensionE9DimType64v", "tensorrt_llm::runtime::ITensor::getDimension::n"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor8getShapeEv", "tensorrt_llm::runtime::ITensor::getShape"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::makeShape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9makeShapeERKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::makeShape::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensoraSERK7ITensor", "tensorrt_llm::runtime::ITensor::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", "tensorrt_llm::runtime::ITensor::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7reshapeERK5Shape", "tensorrt_llm::runtime::ITensor::reshape::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", "tensorrt_llm::runtime::ITensor::resize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6resizeENSt6size_tE", "tensorrt_llm::runtime::ITensor::resize::newSize"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", "tensorrt_llm::runtime::ITensor::shapeEquals"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::T"], [1, 8, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::count"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::count"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::dims"], [1, 4, 1, "_CPPv4I0ENK12tensorrt_llm7runtime7ITensor11shapeEqualsEbPK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::dims"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor11shapeEqualsEbRK5ShapePK1T10SizeType32", "tensorrt_llm::runtime::ITensor::shapeEquals::lhs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::lhs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::other"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7ITensor11shapeEqualsERKNSt16initializer_listI10SizeType32EE", "tensorrt_llm::runtime::ITensor::shapeEquals::other"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor11shapeEqualsERK5ShapeRK5Shape", "tensorrt_llm::runtime::ITensor::shapeEquals::rhs"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::TConstPtr"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offset"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::offsetDims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::size"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor5sliceE14UniqueConstPtrRR9TConstPtrRKNSt16initializer_listI9DimType64EENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrNSt6size_tENSt6size_tE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRK5Shape9DimType64", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor5sliceE9SharedPtrRKNSt16initializer_listI9DimType64EE9DimType64", "tensorrt_llm::runtime::ITensor::slice::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeE10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7squeezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::squeeze::shape"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", "tensorrt_llm::runtime::ITensor::strides"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor7stridesERK5Shape", "tensorrt_llm::runtime::ITensor::strides::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", "tensorrt_llm::runtime::ITensor::toString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor8toStringERK5Shape", "tensorrt_llm::runtime::ITensor::toString::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeE10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::dim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor9unsqueezeERK5Shape10SizeType32", "tensorrt_llm::runtime::ITensor::unsqueeze::shape"], [1, 3, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", "tensorrt_llm::runtime::ITensor::view"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view"], [1, 8, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::TConstPtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view::buffer"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewEN7IBuffer9SharedPtrERK5Shape", "tensorrt_llm::runtime::ITensor::view::dims"], [1, 4, 1, "_CPPv4I0_NSt11enable_if_tINSt10is_const_vI18PointerElementTypeI9TConstPtrEEEiEEEN12tensorrt_llm7runtime7ITensor4viewE14UniqueConstPtrRR9TConstPtrRK5Shape", "tensorrt_llm::runtime::ITensor::view::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4viewE9SharedPtr", "tensorrt_llm::runtime::ITensor::view::tensor"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", "tensorrt_llm::runtime::ITensor::volume"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor6volumeERK5Shape", "tensorrt_llm::runtime::ITensor::volume::dims"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", "tensorrt_llm::runtime::ITensor::volumeNonNegative"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor17volumeNonNegativeERK5Shape", "tensorrt_llm::runtime::ITensor::volumeNonNegative::shape"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::capacity"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::capacity"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::data"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrP1TRK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::shape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::type"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7ITensor4wrapEPvN8nvinfer18DataTypeERK5ShapeNSt6size_tE", "tensorrt_llm::runtime::ITensor::wrap::type"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime7ITensor4wrapE9UniquePtrRNSt6vectorI1TEERK5Shape", "tensorrt_llm::runtime::ITensor::wrap::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7ITensorD0Ev", "tensorrt_llm::runtime::ITensor::~ITensor"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryE", "tensorrt_llm::runtime::IpcMemory"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9BufferPtrE", "tensorrt_llm::runtime::IpcMemory::BufferPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory10FLAGS_SIZEE", "tensorrt_llm::runtime::IpcMemory::FLAGS_SIZE"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERK9IpcMemory", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryERR9IpcMemory", "tensorrt_llm::runtime::IpcMemory::IpcMemory"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::bufferSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::openIpc"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9IpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfigb", "tensorrt_llm::runtime::IpcMemory::IpcMemory::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::bufferSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory17allocateIpcMemoryENSt6size_tERK13BufferManagerRK11WorldConfig", "tensorrt_llm::runtime::IpcMemory::allocateIpcMemory::worldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory16destroyIpcMemoryEv", "tensorrt_llm::runtime::IpcMemory::destroyIpcMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9IpcMemory11getCommPtrsEv", "tensorrt_llm::runtime::IpcMemory::getCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mBufferE", "tensorrt_llm::runtime::IpcMemory::mBuffer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory9mCommPtrsE", "tensorrt_llm::runtime::IpcMemory::mCommPtrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory8mOpenIpcE", "tensorrt_llm::runtime::IpcMemory::mOpenIpc"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemory7mTpRankE", "tensorrt_llm::runtime::IpcMemory::mTpRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERK9IpcMemory", "tensorrt_llm::runtime::IpcMemory::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryaSERR9IpcMemory", "tensorrt_llm::runtime::IpcMemory::operator="], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9IpcMemoryD0Ev", "tensorrt_llm::runtime::IpcMemory::~IpcMemory"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandleE", "tensorrt_llm::runtime::IpcNvlsHandle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle14ipc_uc_handlesE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_handles"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle11ipc_uc_ptrsE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_ptrs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle10ipc_uc_vasE", "tensorrt_llm::runtime::IpcNvlsHandle::ipc_uc_vas"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9mc_handleE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_handle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6mc_ptrE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_ptr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5mc_vaE", "tensorrt_llm::runtime::IpcNvlsHandle::mc_va"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle4sizeE", "tensorrt_llm::runtime::IpcNvlsHandle::size"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle9uc_handleE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_handle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle6uc_ptrE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_ptr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13IpcNvlsHandle5uc_vaE", "tensorrt_llm::runtime::IpcNvlsHandle::uc_va"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffersE", "tensorrt_llm::runtime::LookaheadDecodingBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::maxNumSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers24LookaheadDecodingBuffersE10SizeType3210SizeType32RK13BufferManager", "tensorrt_llm::runtime::LookaheadDecodingBuffers::LookaheadDecodingBuffers::maxTokensPerStep"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers9TensorPtrE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers17generationLengthsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::generationLengths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11packedMasksE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::packedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers11positionIdsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::positionIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime24LookaheadDecodingBuffers15positionOffsetsE", "tensorrt_llm::runtime::LookaheadDecodingBuffers::positionOffsets"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModuleE", "tensorrt_llm::runtime::LookaheadModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleEv", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule15LookaheadModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadModule::LookaheadModule::maxDraftPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime15LookaheadModule18getExecutionConfigEv", "tensorrt_llm::runtime::LookaheadModule::getExecutionConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule16mExecutionConfigE", "tensorrt_llm::runtime::LookaheadModule::mExecutionConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", "tensorrt_llm::runtime::LookaheadModule::setExecutionConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15LookaheadModule18setExecutionConfigERKN8executor23LookaheadDecodingConfigE", "tensorrt_llm::runtime::LookaheadModule::setExecutionConfig::config"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffersE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::decodingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::runtime"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23LookaheadRuntimeBuffersE10SizeType3210SizeType32RK13BufferManagerRK11ModelConfigRK11WorldConfigRKN8executor14DecodingConfigERK11TllmRuntime", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::LookaheadRuntimeBuffers::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorMapE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::TensorMap"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers9TensorPtrE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18batchSlotsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::batchSlotsHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers12cumSumLengthE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::cumSumLength"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers24disableLookaheadDecodingEv", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::disableLookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23enableLookaheadDecodingE10SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::enableLookaheadDecoding::tokensPerStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23generationLengthsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21generationLengthsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers25generationLengthsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::generationLengthsHostCopy"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::inputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::outputBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers18insertInputTensorsER9TensorMapR9TensorMapRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::insertInputTensors::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers14packedMaskHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMaskHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers18packedMaskHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMaskHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17packedMasksDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::packedMasksDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers17positionIdsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15positionIdsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionIdsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionIdsHostCopy"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers21positionOffsetsDeviceE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsDevice"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers19positionOffsetsHostE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsHost"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers23positionOffsetsHostCopyE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::positionOffsetsHostCopy"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::numCtxSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::numGenSequences"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers7reshapeE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::reshape::tokensPerStep"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::decoderLookaheadBuffers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::modelConfig"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::numCtxSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::numGenSequences"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::requestTypes"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::runtime"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::seqSlots"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23LookaheadRuntimeBuffers13setFromInputsE10SizeType3210SizeType32RK7ITensorRK7ITensorRK24LookaheadDecodingBuffersRK11TllmRuntimeRK11ModelConfigRK11WorldConfig", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::setFromInputs::worldConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23LookaheadRuntimeBuffers15useSpecDecodingE", "tensorrt_llm::runtime::LookaheadRuntimeBuffers::useSpecDecoding"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCacheE", "tensorrt_llm::runtime::LoraCache"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::pageManagerConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9LoraCacheERK26LoraCachePageManagerConfigRK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCache::LoraCache::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10TaskIdTypeE", "tensorrt_llm::runtime::LoraCache::TaskIdType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig11adapterSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::adapterSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6inSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::inSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7layerIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::layerId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8moduleIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::moduleId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8numSlotsE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::numSlots"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfigeqERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::operator==::o"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7outSizeE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::outSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig6pageIdE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::pageId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17scalingVecPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::scalingVecPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig7slotIdxE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::slotIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig8toStringEv", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::toString"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig16weightsInPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::weightsInPointer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21TaskLayerModuleConfig17weightsOutPointerE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfig::weightsOutPointer"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache28TaskLayerModuleConfigListPtrE", "tensorrt_llm::runtime::LoraCache::TaskLayerModuleConfigListPtr"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueE", "tensorrt_llm::runtime::LoraCache::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueEv", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::configs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::done"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::inProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::it"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::loadInProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::loaded"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::o"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue9TaskValueERKNSt6vectorINSt6size_tEEERK28TaskLayerModuleConfigListPtrNSt4listI10TaskIdTypeE8iteratorEbbbb", "tensorrt_llm::runtime::LoraCache::TaskValue::TaskValue::pageIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7configsE", "tensorrt_llm::runtime::LoraCache::TaskValue::configs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue4doneE", "tensorrt_llm::runtime::LoraCache::TaskValue::done"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue10inProgressE", "tensorrt_llm::runtime::LoraCache::TaskValue::inProgress"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue2itE", "tensorrt_llm::runtime::LoraCache::TaskValue::it"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue14loadInProgressE", "tensorrt_llm::runtime::LoraCache::TaskValue::loadInProgress"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue6loadedE", "tensorrt_llm::runtime::LoraCache::TaskValue::loaded"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueaSERR9TaskValue", "tensorrt_llm::runtime::LoraCache::TaskValue::operator=::o"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValue7pageIdsE", "tensorrt_llm::runtime::LoraCache::TaskValue::pageIds"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TaskValueD0Ev", "tensorrt_llm::runtime::LoraCache::TaskValue::~TaskValue"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12TaskValuePtrE", "tensorrt_llm::runtime::LoraCache::TaskValuePtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9TensorPtrE", "tensorrt_llm::runtime::LoraCache::TensorPtr"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatusE", "tensorrt_llm::runtime::LoraCache::ValueStatus"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_LOADED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_MISSING"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", "tensorrt_llm::runtime::LoraCache::ValueStatus::kVALUE_STATUS_PROCESSING"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bump"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache4bumpE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bump::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bumpTaskInProgress"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18bumpTaskInProgressE10TaskIdType", "tensorrt_llm::runtime::LoraCache::bumpTaskInProgress::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", "tensorrt_llm::runtime::LoraCache::claimPagesWithEvict"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache19claimPagesWithEvictE10SizeType32", "tensorrt_llm::runtime::LoraCache::claimPagesWithEvict::numPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::deviceCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::markDone"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache8copyTaskE10TaskIdTypeR9LoraCacheb", "tensorrt_llm::runtime::LoraCache::copyTask::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::sourceTaskValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetCache"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetPageIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16copyTaskMapPagesER9TaskValueRK9TaskValueRKNSt6vectorI6size_tEERK9LoraCache", "tensorrt_llm::runtime::LoraCache::copyTaskMapPages::targetTaskValue"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::moduleIdToModel"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::pageIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::pages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::weights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11copyToPagesE9TensorPtr9TensorPtrRK11ModelConfigRK11WorldConfigNSt13unordered_mapI10SizeType3210LoraModuleEERK13BufferManagerRKNSt6vectorI9TensorPtrEERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCache::copyToPages::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", "tensorrt_llm::runtime::LoraCache::determineNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", "tensorrt_llm::runtime::LoraCache::determineNumPages"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE9TensorPtr", "tensorrt_llm::runtime::LoraCache::determineNumPages::config"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache17determineNumPagesE10TaskIdType", "tensorrt_llm::runtime::LoraCache::determineNumPages::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", "tensorrt_llm::runtime::LoraCache::fits"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache4fitsE9TensorPtr", "tensorrt_llm::runtime::LoraCache::fits::config"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", "tensorrt_llm::runtime::LoraCache::get"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3getE10TaskIdType", "tensorrt_llm::runtime::LoraCache::get::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache11getNumPagesEv", "tensorrt_llm::runtime::LoraCache::getNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", "tensorrt_llm::runtime::LoraCache::getPagePtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache10getPagePtrE6size_t", "tensorrt_llm::runtime::LoraCache::getPagePtr::pageId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", "tensorrt_llm::runtime::LoraCache::getStatus"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache9getStatusE10TaskIdType", "tensorrt_llm::runtime::LoraCache::getStatus::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", "tensorrt_llm::runtime::LoraCache::has"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache3hasE10TaskIdType", "tensorrt_llm::runtime::LoraCache::has::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isDone"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache6isDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isDone::taskId"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isLoaded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime9LoraCache8isLoadedE10TaskIdType", "tensorrt_llm::runtime::LoraCache::isLoaded::taskId"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus20kVALUE_STATUS_LOADEDE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_LOADED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus21kVALUE_STATUS_MISSINGE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_MISSING"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11ValueStatus24kVALUE_STATUS_PROCESSINGE", "tensorrt_llm::runtime::LoraCache::kVALUE_STATUS_PROCESSING"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::cacheValue"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::taskId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsE10TaskIdType9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::weights"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11loadWeightsER9TaskValue9TensorPtr9TensorPtr", "tensorrt_llm::runtime::LoraCache::loadWeights::weights"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache14mBufferManagerE", "tensorrt_llm::runtime::LoraCache::mBufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache9mCacheMapE", "tensorrt_llm::runtime::LoraCache::mCacheMap"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mCacheMutexE", "tensorrt_llm::runtime::LoraCache::mCacheMutex"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mCachePageManagerE", "tensorrt_llm::runtime::LoraCache::mCachePageManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache21mDeviceBufferManagersE", "tensorrt_llm::runtime::LoraCache::mDeviceBufferManagers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache10mDoneTasksE", "tensorrt_llm::runtime::LoraCache::mDoneTasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache16mInProgressTasksE", "tensorrt_llm::runtime::LoraCache::mInProgressTasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mModelConfigE", "tensorrt_llm::runtime::LoraCache::mModelConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17mModuleIdToModuleE", "tensorrt_llm::runtime::LoraCache::mModuleIdToModule"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache18mPageManagerConfigE", "tensorrt_llm::runtime::LoraCache::mPageManagerConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11mPagesMutexE", "tensorrt_llm::runtime::LoraCache::mPagesMutex"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12mWorldConfigE", "tensorrt_llm::runtime::LoraCache::mWorldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache11markAllDoneEv", "tensorrt_llm::runtime::LoraCache::markAllDone"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::markTaskDone"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache12markTaskDoneE10TaskIdType", "tensorrt_llm::runtime::LoraCache::markTaskDone::taskId"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::config"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::load"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::taskId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache3putE10TaskIdType9TensorPtr9TensorPtrb", "tensorrt_llm::runtime::LoraCache::put::weights"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::tpRank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9LoraCache17splitTransposeCpuER7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpu::tpSize"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::input"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::output"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::tpRank"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime9LoraCache22splitTransposeCpuInnerEvR7ITensorRK7ITensor10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCache::splitTransposeCpuInner::tpSize"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionE", "tensorrt_llm::runtime::LoraCacheFullException"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraCacheFullException::LoraCacheFullException"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullException22LoraCacheFullExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraCacheFullException::LoraCacheFullException::msg"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime22LoraCacheFullExceptionD0Ev", "tensorrt_llm::runtime::LoraCacheFullException::~LoraCacheFullException"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManagerE", "tensorrt_llm::runtime::LoraCachePageManager"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager20LoraCachePageManagerERK26LoraCachePageManagerConfigRK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::LoraCachePageManager::config"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager9TensorPtrE", "tensorrt_llm::runtime::LoraCachePageManager::TensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::blockPtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager8blockPtrE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::blockPtr::blockIdx"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::claimPages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10claimPagesE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManager::claimPages::numPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::initialize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager10initializeERK13BufferManager", "tensorrt_llm::runtime::LoraCachePageManager::initialize::bufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager7mConfigE", "tensorrt_llm::runtime::LoraCachePageManager::mConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12mFreePageIdsE", "tensorrt_llm::runtime::LoraCachePageManager::mFreePageIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mIsPageFreeE", "tensorrt_llm::runtime::LoraCachePageManager::mIsPageFree"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager11mPageBlocksE", "tensorrt_llm::runtime::LoraCachePageManager::mPageBlocks"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::mutablePagePtr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager14mutablePagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::mutablePagePtr::pageIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager17numAvailablePagesEv", "tensorrt_llm::runtime::LoraCachePageManager::numAvailablePages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::pagePtr"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime20LoraCachePageManager7pagePtrENSt6size_tE", "tensorrt_llm::runtime::LoraCachePageManager::pagePtr::pageIdx"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCachePageManager::releasePages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20LoraCachePageManager12releasePagesERKNSt6vectorINSt6size_tEEE", "tensorrt_llm::runtime::LoraCachePageManager::releasePages::pages"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfigE", "tensorrt_llm::runtime::LoraCachePageManagerConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::dType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::maxPagesPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::memType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::numCopyStreams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::pageWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::slotsPerPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig26LoraCachePageManagerConfigEN7runtime10MemoryTypeEN8nvinfer18DataTypeE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::LoraCachePageManagerConfig::totalNumPages"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig11getDataTypeEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getInitToZeroEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getInitToZero"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig19getMaxPagesPerBlockEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getMaxPagesPerBlock"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig13getMemoryTypeEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getMemoryType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig17getNumCopyStreamsEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getNumCopyStreams"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig12getPageWidthEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getPageWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig15getSlotsPerPageEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getSlotsPerPage"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime26LoraCachePageManagerConfig16getTotalNumPagesEv", "tensorrt_llm::runtime::LoraCachePageManagerConfig::getTotalNumPages"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig9mDataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mInitToZeroE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mInitToZero"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17mMaxPagesPerBlockE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mMaxPagesPerBlock"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11mMemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mMemoryType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15mNumCopyStreamsE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mNumCopyStreams"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig10mPageWidthE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mPageWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13mSlotsPerPageE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mSlotsPerPage"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig14mTotalNumPagesE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::mTotalNumPages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setDataType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig11setDataTypeERKN8nvinfer18DataTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setDataType::dtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setInitToZero"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setInitToZeroEb", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setInitToZero::initToZero"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMaxPagesPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig19setMaxPagesPerBlockERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMaxPagesPerBlock::maxPagesPerBlock"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMemoryType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig13setMemoryTypeERKN7runtime10MemoryTypeE", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setMemoryType::memoryType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setNumCopyStreams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig17setNumCopyStreamsE10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setNumCopyStreams::numCopyStreams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setPageWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig12setPageWidthERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setPageWidth::pageWidth"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setSlotsPerPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setSlotsPerPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setSlotsPerPage::slotsPerPage"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setTotalNumPage"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime26LoraCachePageManagerConfig15setTotalNumPageERK10SizeType32", "tensorrt_llm::runtime::LoraCachePageManagerConfig::setTotalNumPage::totalNumPages"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionE", "tensorrt_llm::runtime::LoraExpectedException"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraExpectedException::LoraExpectedException"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedException21LoraExpectedExceptionERKNSt6stringE", "tensorrt_llm::runtime::LoraExpectedException::LoraExpectedException::msg"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime21LoraExpectedExceptionD0Ev", "tensorrt_llm::runtime::LoraExpectedException::~LoraExpectedException"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleE", "tensorrt_llm::runtime::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleEv", "tensorrt_llm::runtime::LoraModule::LoraModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inDimFirst"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::inTpSplitDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10LoraModule", "tensorrt_llm::runtime::LoraModule::LoraModule::o"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outDimFirst"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::outTpSplitDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10LoraModuleERK10ModuleType10SizeType3210SizeType32bb10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::LoraModule::t"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleTypeE", "tensorrt_llm::runtime::LoraModule::ModuleType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kATTN_DENSEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_DENSE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_KE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_K"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_QE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_Q"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kATTN_QKVE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_QKV"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType7kATTN_VE", "tensorrt_llm::runtime::LoraModule::ModuleType::kATTN_V"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType17kCROSS_ATTN_DENSEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_DENSE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_KE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_K"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_QE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_Q"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType15kCROSS_ATTN_QKVE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_QKV"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType13kCROSS_ATTN_VE", "tensorrt_llm::runtime::LoraModule::ModuleType::kCROSS_ATTN_V"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType8kINVALIDE", "tensorrt_llm::runtime::LoraModule::ModuleType::kINVALID"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_4H_TO_HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_4H_TO_H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMLP_GATEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_GATE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_GATE_UPE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_GATE_UP"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMLP_H_TO_4HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_H_TO_4H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMLP_ROUTERE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMLP_ROUTER"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_4H_TO_HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_4H_TO_H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType9kMOE_GATEE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_GATE"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType12kMOE_H_TO_4HE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_H_TO_4H"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule10ModuleType11kMOE_ROUTERE", "tensorrt_llm::runtime::LoraModule::ModuleType::kMOE_ROUTER"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule9TensorPtrE", "tensorrt_llm::runtime::LoraModule::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::attentionHeadSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::loraModuleNames"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::mlpHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numAttentionHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numExperts"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::numKvAttentionHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule17createLoraModulesERKNSt6vectorINSt6stringEEE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::createLoraModules::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18flattenedInOutSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::flattenedInOutSize::isDora"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5inDimEv", "tensorrt_llm::runtime::LoraModule::inDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10inDimFirstEv", "tensorrt_llm::runtime::LoraModule::inDimFirst"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::inSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6inSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::inSize::adapterSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12inTpSplitDimEv", "tensorrt_llm::runtime::LoraModule::inTpSplitDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule18localInAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInAdapterSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localInDim"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule10localInDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localInDim::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localInOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInOutSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localInSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localInSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule19localOutAdapterSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutAdapterSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localOutDim"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11localOutDimE10SizeType32", "tensorrt_llm::runtime::LoraModule::localOutDim::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule12localOutSizeE10SizeType3210SizeType32", "tensorrt_llm::runtime::LoraModule::localOutSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize::isDora"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule15localScalesSizeE10SizeType32b", "tensorrt_llm::runtime::LoraModule::localScalesSize::tpSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::adapterSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::isDora"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule14localTotalSizeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::LoraModule::localTotalSize::tpSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule6mInDimE", "tensorrt_llm::runtime::LoraModule::mInDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule11mInDimFirstE", "tensorrt_llm::runtime::LoraModule::mInDimFirst"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule13mInTpSplitDimE", "tensorrt_llm::runtime::LoraModule::mInTpSplitDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule7mOutDimE", "tensorrt_llm::runtime::LoraModule::mOutDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12mOutDimFirstE", "tensorrt_llm::runtime::LoraModule::mOutDimFirst"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule14mOutTpSplitDimE", "tensorrt_llm::runtime::LoraModule::mOutTpSplitDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule5mTypeE", "tensorrt_llm::runtime::LoraModule::mType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule4nameEv", "tensorrt_llm::runtime::LoraModule::name"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", "tensorrt_llm::runtime::LoraModule::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModuleaSERK10LoraModule", "tensorrt_llm::runtime::LoraModule::operator=::o"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule6outDimEv", "tensorrt_llm::runtime::LoraModule::outDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule11outDimFirstEv", "tensorrt_llm::runtime::LoraModule::outDimFirst"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::outSize"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule7outSizeE10SizeType32", "tensorrt_llm::runtime::LoraModule::outSize::adapterSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule13outTpSplitDimEv", "tensorrt_llm::runtime::LoraModule::outTpSplitDim"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", "tensorrt_llm::runtime::LoraModule::toModuleName"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", "tensorrt_llm::runtime::LoraModule::toModuleName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10SizeType32", "tensorrt_llm::runtime::LoraModule::toModuleName::id"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleNameE10ModuleType", "tensorrt_llm::runtime::LoraModule::toModuleName::t"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", "tensorrt_llm::runtime::LoraModule::toModuleType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10LoraModule12toModuleTypeERKNSt11string_viewE", "tensorrt_llm::runtime::LoraModule::toModuleType::name"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime10LoraModule5valueEv", "tensorrt_llm::runtime::LoraModule::value"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14LoraTaskIdTypeE", "tensorrt_llm::runtime::LoraTaskIdType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", "tensorrt_llm::runtime::MPI_group_barrier"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime17MPI_group_barrierENSt3setIiEE", "tensorrt_llm::runtime::MPI_group_barrier::ranks"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModuleE", "tensorrt_llm::runtime::MedusaModule"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule13MedusaChoicesE", "tensorrt_llm::runtime::MedusaModule::MedusaChoices"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleEv", "tensorrt_llm::runtime::MedusaModule::MedusaModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule::maxAcceptedTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule12MedusaModuleE10SizeType3210SizeType32", "tensorrt_llm::runtime::MedusaModule::MedusaModule::maxDraftTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule9TensorPtrE", "tensorrt_llm::runtime::MedusaModule::TensorPtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime12MedusaModule16getMedusaChoicesEv", "tensorrt_llm::runtime::MedusaModule::getMedusaChoices"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime12MedusaModule21mDefaultMedusaChoicesE", "tensorrt_llm::runtime::MedusaModule::mDefaultMedusaChoices"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCountersE", "tensorrt_llm::runtime::MemoryCounters"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8DiffTypeE", "tensorrt_llm::runtime::MemoryCounters::DiffType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters14MemoryCountersEv", "tensorrt_llm::runtime::MemoryCounters::MemoryCounters"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10SizeType32E", "tensorrt_llm::runtime::MemoryCounters::SizeType32"], [1, 3, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::T"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::memoryType"], [1, 4, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters8allocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8allocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::allocate::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString::bytes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString::bytes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE10SizeType32i", "tensorrt_llm::runtime::MemoryCounters::bytesToString::precision"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters13bytesToStringE8DiffTypei", "tensorrt_llm::runtime::MemoryCounters::bytesToString::precision"], [1, 3, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::T"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::memoryType"], [1, 4, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime14MemoryCounters10deallocateEv10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::size"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters10deallocateE10MemoryType10SizeType32", "tensorrt_llm::runtime::MemoryCounters::deallocate::size"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getCpuEv", "tensorrt_llm::runtime::MemoryCounters::getCpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getCpuDiffEv", "tensorrt_llm::runtime::MemoryCounters::getCpuDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getGpuEv", "tensorrt_llm::runtime::MemoryCounters::getGpu"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getGpuDiffEv", "tensorrt_llm::runtime::MemoryCounters::getGpuDiff"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11getInstanceEv", "tensorrt_llm::runtime::MemoryCounters::getInstance"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters9getPinnedEv", "tensorrt_llm::runtime::MemoryCounters::getPinned"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedDiffEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters13getPinnedPoolEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedPool"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters17getPinnedPoolDiffEv", "tensorrt_llm::runtime::MemoryCounters::getPinnedPoolDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters6getUVMEv", "tensorrt_llm::runtime::MemoryCounters::getUVM"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters10getUVMDiffEv", "tensorrt_llm::runtime::MemoryCounters::getUVMDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mCpuE", "tensorrt_llm::runtime::MemoryCounters::mCpu"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mCpuDiffE", "tensorrt_llm::runtime::MemoryCounters::mCpuDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mGpuE", "tensorrt_llm::runtime::MemoryCounters::mGpu"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mGpuDiffE", "tensorrt_llm::runtime::MemoryCounters::mGpuDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters7mPinnedE", "tensorrt_llm::runtime::MemoryCounters::mPinned"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedDiffE", "tensorrt_llm::runtime::MemoryCounters::mPinnedDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters11mPinnedPoolE", "tensorrt_llm::runtime::MemoryCounters::mPinnedPool"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters15mPinnedPoolDiffE", "tensorrt_llm::runtime::MemoryCounters::mPinnedPoolDiff"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters4mUVME", "tensorrt_llm::runtime::MemoryCounters::mUVM"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14MemoryCounters8mUVMDiffE", "tensorrt_llm::runtime::MemoryCounters::mUVMDiff"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14MemoryCounters8toStringEv", "tensorrt_llm::runtime::MemoryCounters::toString"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryTypeE", "tensorrt_llm::runtime::MemoryType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kCPUE", "tensorrt_llm::runtime::MemoryType::kCPU"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kGPUE", "tensorrt_llm::runtime::MemoryType::kGPU"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType7kPINNEDE", "tensorrt_llm::runtime::MemoryType::kPINNED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType11kPINNEDPOOLE", "tensorrt_llm::runtime::MemoryType::kPINNEDPOOL"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime10MemoryType4kUVME", "tensorrt_llm::runtime::MemoryType::kUVM"], [1, 2, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", "tensorrt_llm::runtime::MemoryTypeString"], [1, 8, 1, "_CPPv4I_10MemoryTypeEN12tensorrt_llm7runtime16MemoryTypeStringE", "tensorrt_llm::runtime::MemoryTypeString::T"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kCPU>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kCPUEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kCPU>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kGPU>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kGPUEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kGPU>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNED>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType7kPINNEDEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNED>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNEDPOOL>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType11kPINNEDPOOLEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kPINNEDPOOL>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEEE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kUVM>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime16MemoryTypeStringIN10MemoryType4kUVMEE5valueE", "tensorrt_llm::runtime::MemoryTypeString<MemoryType::kUVM>::value"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfigE", "tensorrt_llm::runtime::ModelConfig"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheTypeE", "tensorrt_llm::runtime::ModelConfig::KVCacheType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType11kCONTINUOUSE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kCONTINUOUS"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType9kDISABLEDE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kDISABLED"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11KVCacheType6kPAGEDE", "tensorrt_llm::runtime::ModelConfig::KVCacheType::kPAGED"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", "tensorrt_llm::runtime::ModelConfig::KVCacheTypeFromString"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21KVCacheTypeFromStringENSt6stringE", "tensorrt_llm::runtime::ModelConfig::KVCacheTypeFromString::value"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerTypeE", "tensorrt_llm::runtime::ModelConfig::LayerType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kATTENTIONE", "tensorrt_llm::runtime::ModelConfig::LayerType::kATTENTION"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType7kLINEARE", "tensorrt_llm::runtime::ModelConfig::LayerType::kLINEAR"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType5kNOOPE", "tensorrt_llm::runtime::ModelConfig::LayerType::kNOOP"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9LayerType10kRECURRENTE", "tensorrt_llm::runtime::ModelConfig::LayerType::kRECURRENT"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsTypeE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType9kDisabledE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType::kDisabled"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17ManageWeightsType8kEnabledE", "tensorrt_llm::runtime::ModelConfig::ManageWeightsType::kEnabled"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::hiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbAttentionLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::nbRnnLayers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11ModelConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType3210SizeType32N8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::ModelConfig::vocabSize"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariantE", "tensorrt_llm::runtime::ModelConfig::ModelVariant"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant8kChatGlmE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kChatGlm"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant7kEncDecE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kEncDec"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGlmE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kGlm"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant4kGptE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kGpt"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant6kMambaE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kMamba"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12ModelVariant15kRecurrentGemmaE", "tensorrt_llm::runtime::ModelConfig::ModelVariant::kRecurrentGemma"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfigE", "tensorrt_llm::runtime::ModelConfig::RnnConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig10convKernelE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::convKernel"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig14rnnConvDimSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnConvDimSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig11rnnHeadSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnHeadSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig13rnnHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::rnnHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9RnnConfig9stateSizeE", "tensorrt_llm::runtime::ModelConfig::RnnConfig::stateSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeContextLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEv", "tensorrt_llm::runtime::ModelConfig::computeContextLogits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20computeContextLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeContextLogits::computeContextLogits"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEv", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23computeGenerationLogitsEb", "tensorrt_llm::runtime::ModelConfig::computeGenerationLogits::computeGenerationLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::layerType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16countLocalLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLocalLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::layerType"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20countLowerRankLayersE9LayerType10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::countLowerRankLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig32disableSeamlessLookaheadDecodingEv", "tensorrt_llm::runtime::ModelConfig::disableSeamlessLookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", "tensorrt_llm::runtime::ModelConfig::enableSeamlessLookaheadDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31enableSeamlessLookaheadDecodingE10SizeType32", "tensorrt_llm::runtime::ModelConfig::enableSeamlessLookaheadDecoding::maxDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getContextFMHAEv", "tensorrt_llm::runtime::ModelConfig::getContextFMHA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getDataTypeEv", "tensorrt_llm::runtime::ModelConfig::getDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getEncoderHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getEncoderHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getFirstLocalLayerE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getFirstLocalLayer"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getFirstLocalLayerE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getFirstLocalLayer::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getFirstLocalLayerE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getFirstLocalLayer::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getGemmAllReduceDtypeEv", "tensorrt_llm::runtime::ModelConfig::getGemmAllReduceDtype"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getKVCacheTypeEv", "tensorrt_llm::runtime::ModelConfig::getKVCacheType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getKvDataTypeEv", "tensorrt_llm::runtime::ModelConfig::getKvDataType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13getLayerTypesEv", "tensorrt_llm::runtime::ModelConfig::getLayerTypes"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLogitsDtypeEv", "tensorrt_llm::runtime::ModelConfig::getLogitsDtype"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getLoraModulesEv", "tensorrt_llm::runtime::ModelConfig::getLoraModules"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getManageWeightsTypeEv", "tensorrt_llm::runtime::ModelConfig::getManageWeightsType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBatchSizeEv", "tensorrt_llm::runtime::ModelConfig::getMaxBatchSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxBeamWidthEv", "tensorrt_llm::runtime::ModelConfig::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig25getMaxDecodingDraftTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxDecodingDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getMaxDecodingTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxDecodingTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMaxEncoderLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxEncoderLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxInputLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxInputLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getMaxLoraRankEv", "tensorrt_llm::runtime::ModelConfig::getMaxLoraRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getMaxNumTokensEv", "tensorrt_llm::runtime::ModelConfig::getMaxNumTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24getMaxPositionEmbeddingsEv", "tensorrt_llm::runtime::ModelConfig::getMaxPositionEmbeddings"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig30getMaxPromptEmbeddingTableSizeEv", "tensorrt_llm::runtime::ModelConfig::getMaxPromptEmbeddingTableSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getMaxSequenceLenEv", "tensorrt_llm::runtime::ModelConfig::getMaxSequenceLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16getMlpHiddenSizeEv", "tensorrt_llm::runtime::ModelConfig::getMlpHiddenSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getModelNameEv", "tensorrt_llm::runtime::ModelConfig::getModelName"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getModelVariantEv", "tensorrt_llm::runtime::ModelConfig::getModelVariant"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20getNbAttentionLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbAttentionLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10getNbHeadsEv", "tensorrt_llm::runtime::ModelConfig::getNbHeads"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbKvHeads"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbKvHeads::layerIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig11getNbLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getNbRnnLayersE10SizeType3210SizeType32", "tensorrt_llm::runtime::ModelConfig::getNbRnnLayers::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getNumKvHeadsPerLayerEv", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayer"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::isCrossAttention"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getNumKvHeadsPerLayerLocalRangeE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getNumKvHeadsPerLayerLocalRange::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15getNumLanguagesEv", "tensorrt_llm::runtime::ModelConfig::getNumLanguages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig25getOptProfilesSplitPointsEv", "tensorrt_llm::runtime::ModelConfig::getOptProfilesSplitPoints"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19getPagedContextFMHAEv", "tensorrt_llm::runtime::ModelConfig::getPagedContextFMHA"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getPpReduceScatterEv", "tensorrt_llm::runtime::ModelConfig::getPpReduceScatter"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getQuantModeEv", "tensorrt_llm::runtime::ModelConfig::getQuantMode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getRnnConfigEv", "tensorrt_llm::runtime::ModelConfig::getRnnConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21getRotaryEmbeddingDimEv", "tensorrt_llm::runtime::ModelConfig::getRotaryEmbeddingDim"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14getSizePerHeadEv", "tensorrt_llm::runtime::ModelConfig::getSizePerHead"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig26getSpeculativeDecodingModeEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingMode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28getSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModulePtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig31getSpeculativeDecodingModulePtrEv", "tensorrt_llm::runtime::ModelConfig::getSpeculativeDecodingModulePtr"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::isCrossAttention"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::pipelineParallelism"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getSumLocalKvHeadsE10SizeType3210SizeType32b", "tensorrt_llm::runtime::ModelConfig::getSumLocalKvHeads::pipelineParallelismRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17getTokensPerBlockEv", "tensorrt_llm::runtime::ModelConfig::getTokensPerBlock"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12getVocabSizeEv", "tensorrt_llm::runtime::ModelConfig::getVocabSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getVocabSizePadded"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18getVocabSizePaddedE10SizeType32", "tensorrt_llm::runtime::ModelConfig::getVocabSizePadded::worldSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12hasRnnConfigEv", "tensorrt_llm::runtime::ModelConfig::hasRnnConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig28hasSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::hasSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19isContinuousKVCacheEv", "tensorrt_llm::runtime::ModelConfig::isContinuousKVCache"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig16isKVCacheEnabledEv", "tensorrt_llm::runtime::ModelConfig::isKVCacheEnabled"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig12isMultiModalEv", "tensorrt_llm::runtime::ModelConfig::isMultiModal"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14isPagedKVCacheEv", "tensorrt_llm::runtime::ModelConfig::isPagedKVCache"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig10isRnnBasedEv", "tensorrt_llm::runtime::ModelConfig::isRnnBased"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18isTransformerBasedEv", "tensorrt_llm::runtime::ModelConfig::isTransformerBased"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig9isWhisperEv", "tensorrt_llm::runtime::ModelConfig::isWhisper"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig29kDEFAULT_NUM_TOKENS_PER_BLOCKE", "tensorrt_llm::runtime::ModelConfig::kDEFAULT_NUM_TOKENS_PER_BLOCK"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26kOPT_PROFILES_SPLIT_POINTSE", "tensorrt_llm::runtime::ModelConfig::kOPT_PROFILES_SPLIT_POINTS"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mComputeContextLogitsE", "tensorrt_llm::runtime::ModelConfig::mComputeContextLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mComputeGenerationLogitsE", "tensorrt_llm::runtime::ModelConfig::mComputeGenerationLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mContextFMHAE", "tensorrt_llm::runtime::ModelConfig::mContextFMHA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mDataTypeE", "tensorrt_llm::runtime::ModelConfig::mDataType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mEncoderHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mEncoderHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mGemmAllReduceDtypeE", "tensorrt_llm::runtime::ModelConfig::mGemmAllReduceDtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mInputPackedE", "tensorrt_llm::runtime::ModelConfig::mInputPacked"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mKVCacheTypeE", "tensorrt_llm::runtime::ModelConfig::mKVCacheType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mLayerTypesE", "tensorrt_llm::runtime::ModelConfig::mLayerTypes"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLogitsDtypeE", "tensorrt_llm::runtime::ModelConfig::mLogitsDtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mLoraModulesE", "tensorrt_llm::runtime::ModelConfig::mLoraModules"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mManageWeightsTypeE", "tensorrt_llm::runtime::ModelConfig::mManageWeightsType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBatchSizeE", "tensorrt_llm::runtime::ModelConfig::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxBeamWidthE", "tensorrt_llm::runtime::ModelConfig::mMaxBeamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMaxEncoderLenE", "tensorrt_llm::runtime::ModelConfig::mMaxEncoderLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxInputLenE", "tensorrt_llm::runtime::ModelConfig::mMaxInputLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mMaxLoraRankE", "tensorrt_llm::runtime::ModelConfig::mMaxLoraRank"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mMaxNumTokensE", "tensorrt_llm::runtime::ModelConfig::mMaxNumTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mMaxPositionEmbeddingsE", "tensorrt_llm::runtime::ModelConfig::mMaxPositionEmbeddings"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mMaxPromptEmbeddingTableSizeE", "tensorrt_llm::runtime::ModelConfig::mMaxPromptEmbeddingTableSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mMaxSequenceLenE", "tensorrt_llm::runtime::ModelConfig::mMaxSequenceLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mMlpHiddenSizeE", "tensorrt_llm::runtime::ModelConfig::mMlpHiddenSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mModelNameE", "tensorrt_llm::runtime::ModelConfig::mModelName"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mModelVariantE", "tensorrt_llm::runtime::ModelConfig::mModelVariant"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mNbAttentionLayersE", "tensorrt_llm::runtime::ModelConfig::mNbAttentionLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig8mNbHeadsE", "tensorrt_llm::runtime::ModelConfig::mNbHeads"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mNbLayersE", "tensorrt_llm::runtime::ModelConfig::mNbLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mNbRnnLayersE", "tensorrt_llm::runtime::ModelConfig::mNbRnnLayers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28mNumKvHeadsPerAttentionLayerE", "tensorrt_llm::runtime::ModelConfig::mNumKvHeadsPerAttentionLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig33mNumKvHeadsPerCrossAttentionLayerE", "tensorrt_llm::runtime::ModelConfig::mNumKvHeadsPerCrossAttentionLayer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13mNumLanguagesE", "tensorrt_llm::runtime::ModelConfig::mNumLanguages"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17mPagedContextFMHAE", "tensorrt_llm::runtime::ModelConfig::mPagedContextFMHA"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11mPagedStateE", "tensorrt_llm::runtime::ModelConfig::mPagedState"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16mPpReduceScatterE", "tensorrt_llm::runtime::ModelConfig::mPpReduceScatter"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mQuantModeE", "tensorrt_llm::runtime::ModelConfig::mQuantMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mRnnConfigE", "tensorrt_llm::runtime::ModelConfig::mRnnConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19mRotaryEmbeddingDimE", "tensorrt_llm::runtime::ModelConfig::mRotaryEmbeddingDim"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12mSizePerHeadE", "tensorrt_llm::runtime::ModelConfig::mSizePerHead"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20mSkipCrossAttnBlocksE", "tensorrt_llm::runtime::ModelConfig::mSkipCrossAttnBlocks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24mSpeculativeDecodingModeE", "tensorrt_llm::runtime::ModelConfig::mSpeculativeDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26mSpeculativeDecodingModuleE", "tensorrt_llm::runtime::ModelConfig::mSpeculativeDecodingModule"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15mTokensPerBlockE", "tensorrt_llm::runtime::ModelConfig::mTokensPerBlock"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseCrossAttentionE", "tensorrt_llm::runtime::ModelConfig::mUseCrossAttention"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23mUseGemmAllReducePluginE", "tensorrt_llm::runtime::ModelConfig::mUseGemmAllReducePlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseGptAttentionPluginE", "tensorrt_llm::runtime::ModelConfig::mUseGptAttentionPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14mUseLoraPluginE", "tensorrt_llm::runtime::ModelConfig::mUseLoraPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUseMambaConv1dPluginE", "tensorrt_llm::runtime::ModelConfig::mUseMambaConv1dPlugin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig9mUseMropeE", "tensorrt_llm::runtime::ModelConfig::mUseMrope"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21mUsePositionEmbeddingE", "tensorrt_llm::runtime::ModelConfig::mUsePositionEmbedding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18mUseShapeInferenceE", "tensorrt_llm::runtime::ModelConfig::mUseShapeInference"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22mUseTokenTypeEmbeddingE", "tensorrt_llm::runtime::ModelConfig::mUseTokenTypeEmbedding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig10mVocabSizeE", "tensorrt_llm::runtime::ModelConfig::mVocabSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30resetSpeculativeDecodingModuleEv", "tensorrt_llm::runtime::ModelConfig::resetSpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setContextFMHA"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setContextFMHA::contextFMHA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setEncoderHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setEncoderHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setEncoderHiddenSize::encoderHiddenSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setGemmAllReduceDtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setGemmAllReduceDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setGemmAllReduceDtype::inputDtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", "tensorrt_llm::runtime::ModelConfig::setKVCacheType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setKVCacheTypeE11KVCacheType", "tensorrt_llm::runtime::ModelConfig::setKVCacheType::kvCacheType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", "tensorrt_llm::runtime::ModelConfig::setLayerTypes"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13setLayerTypesERKNSt6vectorI9LayerTypeEE", "tensorrt_llm::runtime::ModelConfig::setLayerTypes::layerTypes"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setLogitsDtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLogitsDtypeEN8nvinfer18DataTypeE", "tensorrt_llm::runtime::ModelConfig::setLogitsDtype::inputDtype"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", "tensorrt_llm::runtime::ModelConfig::setLoraModules"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setLoraModulesERKNSt6vectorI10LoraModuleEE", "tensorrt_llm::runtime::ModelConfig::setLoraModules::loraModules"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", "tensorrt_llm::runtime::ModelConfig::setManageWeightsType"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setManageWeightsTypeEK17ManageWeightsType", "tensorrt_llm::runtime::ModelConfig::setManageWeightsType::manageWeightType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBatchSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBatchSize::maxBatchSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxBeamWidthE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxBeamWidth::maxBeamWidth"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxEncoderLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMaxEncoderLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxEncoderLen::maxEncoderLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxInputLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxInputLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxInputLen::maxInputLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxLoraRank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setMaxLoraRankE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxLoraRank::maxLoraRank"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setMaxNumTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setMaxNumTokensENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setMaxNumTokens::maxNumTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPositionEmbeddings"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setMaxPositionEmbeddingsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPositionEmbeddings::maxPositionEmbeddings"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPromptEmbeddingTableSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig30setMaxPromptEmbeddingTableSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxPromptEmbeddingTableSize::maxPromptEmbeddingTableSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxSequenceLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setMaxSequenceLenE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMaxSequenceLen::maxSequenceLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMlpHiddenSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig16setMlpHiddenSizeE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setMlpHiddenSize::mlpHiddenSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", "tensorrt_llm::runtime::ModelConfig::setModelName"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setModelNameERKNSt6stringE", "tensorrt_llm::runtime::ModelConfig::setModelName::modelName"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", "tensorrt_llm::runtime::ModelConfig::setModelVariant"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setModelVariantE12ModelVariant", "tensorrt_llm::runtime::ModelConfig::setModelVariant::modelVariant"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbCrossKvHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setNbCrossKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbCrossKvHeads::nbKvHeads"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbKvHeads"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setNbKvHeadsE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setNbKvHeads::nbKvHeads"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerCrossLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setNumKvHeadsPerCrossLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerCrossLayer::headsPerLayer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerLayer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setNumKvHeadsPerLayerERKNSt6vectorI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumKvHeadsPerLayer::headsPerLayer"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumLanguages"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig15setNumLanguagesENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::ModelConfig::setNumLanguages::numLanguages"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setPagedContextFMHA"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig19setPagedContextFMHAEb", "tensorrt_llm::runtime::ModelConfig::setPagedContextFMHA::pagedContextFMHA"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", "tensorrt_llm::runtime::ModelConfig::setPpReduceScatter"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig18setPpReduceScatterEb", "tensorrt_llm::runtime::ModelConfig::setPpReduceScatter::ppReduceScatter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", "tensorrt_llm::runtime::ModelConfig::setQuantMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setQuantModeEN6common9QuantModeE", "tensorrt_llm::runtime::ModelConfig::setQuantMode::QuantMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", "tensorrt_llm::runtime::ModelConfig::setRnnConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig12setRnnConfigERK9RnnConfig", "tensorrt_llm::runtime::ModelConfig::setRnnConfig::rnnConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setRotaryEmbeddingDim"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21setRotaryEmbeddingDimE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setRotaryEmbeddingDim::rotaryEmbeddingDim"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setSizePerHead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14setSizePerHeadE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setSizePerHead::sizePerHead"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", "tensorrt_llm::runtime::ModelConfig::setSkipCrossAttnBlocks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22setSkipCrossAttnBlocksEb", "tensorrt_llm::runtime::ModelConfig::setSkipCrossAttnBlocks::skipCrossAttnBlocks"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig26setSpeculativeDecodingModeE23SpeculativeDecodingMode", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingMode::mode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig28setSpeculativeDecodingModuleERKNSt10shared_ptrI25SpeculativeDecodingModuleEE", "tensorrt_llm::runtime::ModelConfig::setSpeculativeDecodingModule::speculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setTokensPerBlock"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig17setTokensPerBlockE10SizeType32", "tensorrt_llm::runtime::ModelConfig::setTokensPerBlock::TokensPerBlock"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", "tensorrt_llm::runtime::ModelConfig::setUseCrossAttention"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseCrossAttentionEb", "tensorrt_llm::runtime::ModelConfig::setUseCrossAttention::useCrossAttention"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", "tensorrt_llm::runtime::ModelConfig::setUseMrope"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig11setUseMropeEb", "tensorrt_llm::runtime::ModelConfig::setUseMrope::useMrope"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUsePositionEmbedding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig23setUsePositionEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUsePositionEmbedding::usePositionEmbedding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", "tensorrt_llm::runtime::ModelConfig::setUseShapeInference"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20setUseShapeInferenceEb", "tensorrt_llm::runtime::ModelConfig::setUseShapeInference::useShapeInference"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUseTokenTypeEmbedding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig24setUseTokenTypeEmbeddingEb", "tensorrt_llm::runtime::ModelConfig::setUseTokenTypeEmbedding::useTokenTypeEmbedding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig19skipCrossAttnBlocksEv", "tensorrt_llm::runtime::ModelConfig::skipCrossAttnBlocks"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig24supportsInflightBatchingEv", "tensorrt_llm::runtime::ModelConfig::supportsInflightBatching"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useCrossAttentionEv", "tensorrt_llm::runtime::ModelConfig::useCrossAttention"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEv", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig22useGemmAllReducePluginEb", "tensorrt_llm::runtime::ModelConfig::useGemmAllReducePlugin::useGemmAllReducePlugin"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEv", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig21useGptAttentionPluginEb", "tensorrt_llm::runtime::ModelConfig::useGptAttentionPlugin::useGptAttentionPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig18useLanguageAdapterEv", "tensorrt_llm::runtime::ModelConfig::useLanguageAdapter"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13useLoraPluginEv", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13useLoraPluginEb", "tensorrt_llm::runtime::ModelConfig::useLoraPlugin::useLoraPlugin"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEv", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig20useMambaConv1dPluginEb", "tensorrt_llm::runtime::ModelConfig::useMambaConv1dPlugin::useMambaConv1dPlugin"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig8useMropeEv", "tensorrt_llm::runtime::ModelConfig::useMrope"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", "tensorrt_llm::runtime::ModelConfig::usePackedInput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig14usePackedInputEv", "tensorrt_llm::runtime::ModelConfig::usePackedInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig14usePackedInputEb", "tensorrt_llm::runtime::ModelConfig::usePackedInput::inputPacked"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", "tensorrt_llm::runtime::ModelConfig::usePagedState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig13usePagedStateEv", "tensorrt_llm::runtime::ModelConfig::usePagedState"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ModelConfig13usePagedStateEb", "tensorrt_llm::runtime::ModelConfig::usePagedState::pagedState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig20usePositionEmbeddingEv", "tensorrt_llm::runtime::ModelConfig::usePositionEmbedding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig15usePromptTuningEv", "tensorrt_llm::runtime::ModelConfig::usePromptTuning"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig17useShapeInferenceEv", "tensorrt_llm::runtime::ModelConfig::useShapeInference"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11ModelConfig21useTokenTypeEmbeddingEv", "tensorrt_llm::runtime::ModelConfig::useTokenTypeEmbedding"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", "tensorrt_llm::runtime::PointerElementType"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime18PointerElementTypeE", "tensorrt_llm::runtime::PointerElementType::T"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParamsE", "tensorrt_llm::runtime::PromptTuningParams"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::embeddingTable"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::tasks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams18PromptTuningParamsE9TensorPtr9TensorPtr9TensorPtr", "tensorrt_llm::runtime::PromptTuningParams::PromptTuningParams::vocabSize"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams10SizeType32E", "tensorrt_llm::runtime::PromptTuningParams::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams9TensorPtrE", "tensorrt_llm::runtime::PromptTuningParams::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::batchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::manager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::numContextRequests"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::packedInput"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::reqBeamWidths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::reqPromptLengths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime18PromptTuningParams15fillTasksTensorE9TensorPtr10SizeType3210SizeType32RKNSt6vectorI10SizeType32EERKNSt6vectorI10SizeType32EERK13BufferManagerb", "tensorrt_llm::runtime::PromptTuningParams::fillTasksTensor::tasksHost"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngineE", "tensorrt_llm::runtime::RawEngine"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", "tensorrt_llm::runtime::RawEngine::AddressWithSize"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", "tensorrt_llm::runtime::RawEngine::FilePath"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", "tensorrt_llm::runtime::RawEngine::HostMemory"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineAddr"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKN8nvinfer111IHostMemoryE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineBuffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::RawEngine::enginePath"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine9RawEngineEPKvNSt6size_tE", "tensorrt_llm::runtime::RawEngine::RawEngine::engineSize"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4TypeE", "tensorrt_llm::runtime::RawEngine::Type"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type15AddressWithSizeE", "tensorrt_llm::runtime::RawEngine::Type::AddressWithSize"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type8FilePathE", "tensorrt_llm::runtime::RawEngine::Type::FilePath"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine4Type10HostMemoryE", "tensorrt_llm::runtime::RawEngine::Type::HostMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getAddressEv", "tensorrt_llm::runtime::RawEngine::getAddress"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine13getHostMemoryEv", "tensorrt_llm::runtime::RawEngine::getHostMemory"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine23getManagedWeightsMapOptEv", "tensorrt_llm::runtime::RawEngine::getManagedWeightsMapOpt"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getPathEv", "tensorrt_llm::runtime::RawEngine::getPath"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine10getPathOptEv", "tensorrt_llm::runtime::RawEngine::getPathOpt"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getSizeEv", "tensorrt_llm::runtime::RawEngine::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime9RawEngine7getTypeEv", "tensorrt_llm::runtime::RawEngine::getType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineAddrE", "tensorrt_llm::runtime::RawEngine::mEngineAddr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine13mEngineBufferE", "tensorrt_llm::runtime::RawEngine::mEngineBuffer"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEnginePathE", "tensorrt_llm::runtime::RawEngine::mEnginePath"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine11mEngineSizeE", "tensorrt_llm::runtime::RawEngine::mEngineSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine18mManagedWeightsMapE", "tensorrt_llm::runtime::RawEngine::mManagedWeightsMap"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine5mTypeE", "tensorrt_llm::runtime::RawEngine::mType"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", "tensorrt_llm::runtime::RawEngine::setManagedWeightsMap"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine20setManagedWeightsMapENSt3mapINSt6stringEN12tensorrt_llm8executor6TensorEEE", "tensorrt_llm::runtime::RawEngine::setManagedWeightsMap::managedWeightsMap"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::setPath"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9RawEngine7setPathENSt10filesystem4pathE", "tensorrt_llm::runtime::RawEngine::setPath::enginePath"], [1, 6, 1, "_CPPv4N12tensorrt_llm7runtime11RequestTypeE", "tensorrt_llm::runtime::RequestType"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11RequestType8kCONTEXTE", "tensorrt_llm::runtime::RequestType::kCONTEXT"], [1, 7, 1, "_CPPv4N12tensorrt_llm7runtime11RequestType11kGENERATIONE", "tensorrt_llm::runtime::RequestType::kGENERATION"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaultsE", "tensorrt_llm::runtime::RuntimeDefaults"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsEv", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults::maxAttentionWindowVec"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15RuntimeDefaultsENSt8optionalINSt6vectorI10SizeType32EEEENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::RuntimeDefaults::RuntimeDefaults::sinkTokenLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults21maxAttentionWindowVecE", "tensorrt_llm::runtime::RuntimeDefaults::maxAttentionWindowVec"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime15RuntimeDefaults15sinkTokenLengthE", "tensorrt_llm::runtime::RuntimeDefaults::sinkTokenLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfigE", "tensorrt_llm::runtime::SamplingConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9FloatTypeE", "tensorrt_llm::runtime::SamplingConfig::FloatType"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", "tensorrt_llm::runtime::SamplingConfig::OptVec"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig6OptVecE", "tensorrt_llm::runtime::SamplingConfig::OptVec::T"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigE10SizeType32", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::beamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKNSt6vectorI14SamplingConfigEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::configs"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::externalDraftTokensConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14SamplingConfigERKN8executor14SamplingConfigERKNSt8optionalIN8executor25ExternalDraftTokensConfigEEE", "tensorrt_llm::runtime::SamplingConfig::SamplingConfig::samplingConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig23beamSearchDiversityRateE", "tensorrt_llm::runtime::SamplingConfig::beamSearchDiversityRate"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9beamWidthE", "tensorrt_llm::runtime::SamplingConfig::beamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14beamWidthArrayE", "tensorrt_llm::runtime::SamplingConfig::beamWidthArray"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11cumLogProbsE", "tensorrt_llm::runtime::SamplingConfig::cumLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig24draftAcceptanceThresholdE", "tensorrt_llm::runtime::SamplingConfig::draftAcceptanceThreshold"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13earlyStoppingE", "tensorrt_llm::runtime::SamplingConfig::earlyStopping"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig16frequencyPenaltyE", "tensorrt_llm::runtime::SamplingConfig::frequencyPenalty"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::accessor"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::configs"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig10fuseValuesE6OptVecI1TERKNSt6vectorI14SamplingConfigEENSt8functionIF6OptVecI1TE6size_tEEE1T", "tensorrt_llm::runtime::SamplingConfig::fuseValues::defaultValue"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig15getMaxBeamWidthEv", "tensorrt_llm::runtime::SamplingConfig::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfig17getNumReturnBeamsEv", "tensorrt_llm::runtime::SamplingConfig::getNumReturnBeams"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig13lengthPenaltyE", "tensorrt_llm::runtime::SamplingConfig::lengthPenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9minLengthE", "tensorrt_llm::runtime::SamplingConfig::minLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4minPE", "tensorrt_llm::runtime::SamplingConfig::minP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17noRepeatNgramSizeE", "tensorrt_llm::runtime::SamplingConfig::noRepeatNgramSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17normalizeLogProbsE", "tensorrt_llm::runtime::SamplingConfig::normalizeLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig18numReturnSequencesE", "tensorrt_llm::runtime::SamplingConfig::numReturnSequences"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::runtime::SamplingConfig::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime14SamplingConfigeqERK14SamplingConfig", "tensorrt_llm::runtime::SamplingConfig::operator==::other"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig19originalTemperatureE", "tensorrt_llm::runtime::SamplingConfig::originalTemperature"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig14outputLogProbsE", "tensorrt_llm::runtime::SamplingConfig::outputLogProbs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15presencePenaltyE", "tensorrt_llm::runtime::SamplingConfig::presencePenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig10randomSeedE", "tensorrt_llm::runtime::SamplingConfig::randomSeed"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig17repetitionPenaltyE", "tensorrt_llm::runtime::SamplingConfig::repetitionPenalty"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig11temperatureE", "tensorrt_llm::runtime::SamplingConfig::temperature"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topKE", "tensorrt_llm::runtime::SamplingConfig::topK"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig15topKMedusaHeadsE", "tensorrt_llm::runtime::SamplingConfig::topKMedusaHeads"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig4topPE", "tensorrt_llm::runtime::SamplingConfig::topP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig9topPDecayE", "tensorrt_llm::runtime::SamplingConfig::topPDecay"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig7topPMinE", "tensorrt_llm::runtime::SamplingConfig::topPMin"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig12topPResetIdsE", "tensorrt_llm::runtime::SamplingConfig::topPResetIds"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::defaultValue"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig16useDefaultValuesEbRK6OptVecI1TE1T", "tensorrt_llm::runtime::SamplingConfig::useDefaultValues::vec"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime14SamplingConfig8validateEv", "tensorrt_llm::runtime::SamplingConfig::validate"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::max"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::min"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::name"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime14SamplingConfig11validateVecEbNSt6stringERK6OptVecI1TE1TNSt8optionalI1TEE", "tensorrt_llm::runtime::SamplingConfig::validateVec::vec"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10SizeType32E", "tensorrt_llm::runtime::SizeType32"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime10SizeType64E", "tensorrt_llm::runtime::SizeType64"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingModeE", "tensorrt_llm::runtime::SpeculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19DraftTokensExternalEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::DraftTokensExternal"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5EagleEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::Eagle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode19ExplicitDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::ExplicitDraftTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode17LookaheadDecodingEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::LookaheadDecoding"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6MedusaEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::Medusa"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode4NoneEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::None"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::SpeculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode23SpeculativeDecodingModeE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::SpeculativeDecodingMode::state"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode14UnderlyingTypeE", "tensorrt_llm::runtime::SpeculativeDecodingMode::UnderlyingType"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::allBitSet"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9allBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::allBitSet::bits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::anyBitSet"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode9anyBitSetE14UnderlyingType", "tensorrt_llm::runtime::SpeculativeDecodingMode::anyBitSet::bits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode14hasDraftLogitsEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::hasDraftLogits"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isDraftTokensExternalEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isDraftTokensExternal"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode7isEagleEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isEagle"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21isExplicitDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isExplicitDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19isLookaheadDecodingEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isLookaheadDecoding"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode8isMedusaEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isMedusa"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode6isNoneEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::isNone"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kDraftTokensExternalE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kDraftTokensExternal"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6kEagleE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kEagle"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode20kExplicitDraftTokensE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kExplicitDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode18kLookaheadDecodingE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kLookaheadDecoding"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode7kMedusaE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kMedusa"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode5kNoneE", "tensorrt_llm::runtime::SpeculativeDecodingMode::kNone"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime23SpeculativeDecodingMode6mStateE", "tensorrt_llm::runtime::SpeculativeDecodingMode::mState"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode20needsDecoderPrologueEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::needsDecoderPrologue"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18needsKVCacheRewindEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::needsKVCacheRewind"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", "tensorrt_llm::runtime::SpeculativeDecodingMode::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingModeeqERK23SpeculativeDecodingMode", "tensorrt_llm::runtime::SpeculativeDecodingMode::operator==::other"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19predictsDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::predictsDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode21requiresAttentionMaskEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::requiresAttentionMask"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode18updatesPositionIdsEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::updatesPositionIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime23SpeculativeDecodingMode19variableDraftLengthEv", "tensorrt_llm::runtime::SpeculativeDecodingMode::variableDraftLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleE", "tensorrt_llm::runtime::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxDecodingDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleE10SizeType3210SizeType3210SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::maxNumPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule25SpeculativeDecodingModuleERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::SpeculativeDecodingModule::o"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule21computeNumPackedMasksEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::computeNumPackedMasks"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule25getMaxDecodingDraftTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDecodingDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule20getMaxDecodingTokensEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDecodingTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule18getMaxDraftPathLenEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxDraftPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule14getMaxNumPathsEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxNumPaths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule13getMaxPathLenEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getMaxPathLen"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime25SpeculativeDecodingModule17getNumPackedMasksEv", "tensorrt_llm::runtime::SpeculativeDecodingModule::getNumPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule23mMaxDecodingDraftTokensE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxDecodingDraftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule16mMaxDraftPathLenE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxDraftPathLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18mMaxNumPackedMasksE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxNumPackedMasks"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule12mMaxNumPathsE", "tensorrt_llm::runtime::SpeculativeDecodingModule::mMaxNumPaths"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::operator="], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleaSERK25SpeculativeDecodingModule", "tensorrt_llm::runtime::SpeculativeDecodingModule::operator=::o"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftPathLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule18setMaxDraftPathLenE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftPathLen::maxDraftPathLen"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule17setMaxDraftTokensE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxDraftTokens::maxDraftTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxNumPaths"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModule14setMaxNumPathsE10SizeType32", "tensorrt_llm::runtime::SpeculativeDecodingModule::setMaxNumPaths::maxNumPaths"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime25SpeculativeDecodingModuleD0Ev", "tensorrt_llm::runtime::SpeculativeDecodingModule::~SpeculativeDecodingModule"], [1, 1, 1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", "tensorrt_llm::runtime::StringPtrMap"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime12StringPtrMapE", "tensorrt_llm::runtime::StringPtrMap::T"], [1, 2, 1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", "tensorrt_llm::runtime::TRTDataType"], [1, 8, 1, "_CPPv4I0_bEN12tensorrt_llm7runtime11TRTDataTypeE", "tensorrt_llm::runtime::TRTDataType::T"], [1, 2, 1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", "tensorrt_llm::runtime::TRTDataType<T*>"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime11TRTDataTypeIP1TEE", "tensorrt_llm::runtime::TRTDataType<T*>::T"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE15kUnderlyingTypeE", "tensorrt_llm::runtime::TRTDataType<T*>::kUnderlyingType"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIP1TE5valueE", "tensorrt_llm::runtime::TRTDataType<T*>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIbEE", "tensorrt_llm::runtime::TRTDataType<bool>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIbE5valueE", "tensorrt_llm::runtime::TRTDataType<bool>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIfEE", "tensorrt_llm::runtime::TRTDataType<float>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIfE5valueE", "tensorrt_llm::runtime::TRTDataType<float>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeI4halfEE", "tensorrt_llm::runtime::TRTDataType<half>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeI4halfE5valueE", "tensorrt_llm::runtime::TRTDataType<half>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEEE", "tensorrt_llm::runtime::TRTDataType<kernels::FinishedState>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels13FinishedStateEE5valueE", "tensorrt_llm::runtime::TRTDataType<kernels::FinishedState>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEEE", "tensorrt_llm::runtime::TRTDataType<kernels::KVCacheIndex>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7kernels12KVCacheIndexEE5valueE", "tensorrt_llm::runtime::TRTDataType<kernels::KVCacheIndex>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEEE", "tensorrt_llm::runtime::TRTDataType<runtime::RequestType>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIN7runtime11RequestTypeEE5valueE", "tensorrt_llm::runtime::TRTDataType<runtime::RequestType>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int32_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int32_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int32_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int64_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7int64_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int64_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEEE", "tensorrt_llm::runtime::TRTDataType<std::int8_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt6int8_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::int8_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint32_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint32_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint32_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint64_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt8uint64_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint64_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEEE", "tensorrt_llm::runtime::TRTDataType<std::uint8_t>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeINSt7uint8_tEE5valueE", "tensorrt_llm::runtime::TRTDataType<std::uint8_t>::value"], [1, 2, 1, "_CPPv4IEN12tensorrt_llm7runtime11TRTDataTypeIPvEE", "tensorrt_llm::runtime::TRTDataType<void*>"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11TRTDataTypeIPvE5valueE", "tensorrt_llm::runtime::TRTDataType<void*>::value"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLoggerE", "tensorrt_llm::runtime::TllmLogger"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8getLevelEv", "tensorrt_llm::runtime::TllmLogger::getLevel"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log::msg"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger3logE8SeverityPKN8nvinfer19AsciiCharE", "tensorrt_llm::runtime::TllmLogger::log::severity"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", "tensorrt_llm::runtime::TllmLogger::setLevel"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime10TllmLogger8setLevelE8Severity", "tensorrt_llm::runtime::TllmLogger::setLevel::level"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16TokenExtraIdTypeE", "tensorrt_llm::runtime::TokenExtraIdType"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime11TokenIdTypeE", "tensorrt_llm::runtime::TokenIdType"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueTokenE", "tensorrt_llm::runtime::UniqueToken"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", "tensorrt_llm::runtime::UniqueToken::operator=="], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11UniqueTokeneqERK11UniqueToken", "tensorrt_llm::runtime::UniqueToken::operator==::other"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken12tokenExtraIdE", "tensorrt_llm::runtime::UniqueToken::tokenExtraId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11UniqueToken7tokenIdE", "tensorrt_llm::runtime::UniqueToken::tokenId"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime16VecTokenExtraIdsE", "tensorrt_llm::runtime::VecTokenExtraIds"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime15VecUniqueTokensE", "tensorrt_llm::runtime::VecUniqueTokens"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfigE", "tensorrt_llm::runtime::WorldConfig"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::deviceIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::enableAttentionDP"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::rank"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig11WorldConfigE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::WorldConfig::tensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17enableAttentionDPEv", "tensorrt_llm::runtime::WorldConfig::enableAttentionDP"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getContextParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getContextParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getContextParallelismEv", "tensorrt_llm::runtime::WorldConfig::getContextParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig9getDeviceEv", "tensorrt_llm::runtime::WorldConfig::getDevice"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getDeviceOf"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getDeviceOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getDeviceOf::rank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig15getGpusPerGroupEv", "tensorrt_llm::runtime::WorldConfig::getGpusPerGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14getGpusPerNodeEv", "tensorrt_llm::runtime::WorldConfig::getGpusPerNode"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getLastRankEv", "tensorrt_llm::runtime::WorldConfig::getLastRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig12getLocalRankEv", "tensorrt_llm::runtime::WorldConfig::getLocalRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig11getNodeRankEv", "tensorrt_llm::runtime::WorldConfig::getNodeRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getNodeRankOf"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig13getNodeRankOfE10SizeType32", "tensorrt_llm::runtime::WorldConfig::getNodeRankOf::rank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig24getPipelineParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig23getPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getPipelineParallelismEv", "tensorrt_llm::runtime::WorldConfig::getPipelineParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getRankEv", "tensorrt_llm::runtime::WorldConfig::getRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig7getSizeEv", "tensorrt_llm::runtime::WorldConfig::getSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig22getTensorParallelGroupEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelGroup"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig21getTensorParallelRankEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig20getTensorParallelismEv", "tensorrt_llm::runtime::WorldConfig::getTensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig17isContextParallelEv", "tensorrt_llm::runtime::WorldConfig::isContextParallel"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isFirstContextParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstContextParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig27isFirstPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig25isFirstTensorParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isFirstTensorParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig26isLastPipelineParallelRankEv", "tensorrt_llm::runtime::WorldConfig::isLastPipelineParallelRank"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig18isPipelineParallelEv", "tensorrt_llm::runtime::WorldConfig::isPipelineParallel"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig16isTensorParallelEv", "tensorrt_llm::runtime::WorldConfig::isTensorParallel"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19kDefaultGpusPerNodeE", "tensorrt_llm::runtime::WorldConfig::kDefaultGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig19mContextParallelismE", "tensorrt_llm::runtime::WorldConfig::mContextParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig10mDeviceIdsE", "tensorrt_llm::runtime::WorldConfig::mDeviceIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mEnableAttentionDPE", "tensorrt_llm::runtime::WorldConfig::mEnableAttentionDP"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig12mGpusPerNodeE", "tensorrt_llm::runtime::WorldConfig::mGpusPerNode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig20mPipelineParallelismE", "tensorrt_llm::runtime::WorldConfig::mPipelineParallelism"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig5mRankE", "tensorrt_llm::runtime::WorldConfig::mRank"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig18mTensorParallelismE", "tensorrt_llm::runtime::WorldConfig::mTensorParallelism"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::contextParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::deviceIds"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::enableAttentionDP"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::gpusPerNode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::pipelineParallelism"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11WorldConfig3mpiE10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EENSt8optionalI10SizeType32EERKNSt8optionalINSt6vectorI10SizeType32EEEEb", "tensorrt_llm::runtime::WorldConfig::mpi::tensorParallelism"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime11WorldConfig14validMpiConfigEv", "tensorrt_llm::runtime::WorldConfig::validMpiConfig"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEP1TR7IBuffer", "tensorrt_llm::runtime::bufferCast::buffer"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime10bufferCastEPK1TRK7IBuffer", "tensorrt_llm::runtime::bufferCast::buffer"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::T"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7IBuffer9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::bufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7IBuffer14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::bufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7IBuffer9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalBufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7IBuffer14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalBufferPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKNSt8optionalIN7ITensor9SharedPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalTensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKNSt8optionalIN7ITensor14SharedConstPtrEEE", "tensorrt_llm::runtime::bufferCastOrNull::optionalTensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEP1TRKN7ITensor9SharedPtrE", "tensorrt_llm::runtime::bufferCastOrNull::tensorPtr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16bufferCastOrNullEPK1TRKN7ITensor14SharedConstPtrE", "tensorrt_llm::runtime::bufferCastOrNull::tensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", "tensorrt_llm::runtime::canAccessPeer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13canAccessPeerERK11WorldConfig", "tensorrt_llm::runtime::canAccessPeer::worldConfig"], [1, 3, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast"], [1, 3, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast"], [1, 8, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::D"], [1, 8, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::T"], [1, 8, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast::T"], [1, 4, 1, "_CPPv4I00EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERRNSt10unique_ptrI1T1DEE", "tensorrt_llm::runtime::constPointerCast::ptr"], [1, 4, 1, "_CPPv4I0EN12tensorrt_llm7runtime16constPointerCastENSt10shared_ptrINSt14remove_const_tI1TEEEERKNSt10shared_ptrI1TEE", "tensorrt_llm::runtime::constPointerCast::ptr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoderE", "tensorrt_llm::runtime::decoder"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoderE", "tensorrt_llm::runtime::decoder"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffersE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::BeamSearchBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers17BeamSearchBuffersERK13BufferManager", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::BeamSearchBuffers::bufferManager"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers15mCumLogProbsTmpE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mCumLogProbsTmp"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7mNumSMsE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mNumSMs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers21mOutputBeamHypothesesE", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::mOutputBeamHypotheses"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder17BeamSearchBuffers7reshapeE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::BeamSearchBuffers::reshape::maxSequenceLength"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderStateE", "tensorrt_llm::runtime::decoder::DecoderState"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState12DecoderStateEN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::DecoderState::dtype"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16DecodingInputPtrE", "tensorrt_llm::runtime::decoder::DecoderState::DecodingInputPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState17DecodingOutputPtrE", "tensorrt_llm::runtime::decoder::DecoderState::DecodingOutputPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13LlmRequestPtrE", "tensorrt_llm::runtime::decoder::DecoderState::LlmRequestPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13RequestVectorE", "tensorrt_llm::runtime::decoder::DecoderState::RequestVector"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState9TensorPtrE", "tensorrt_llm::runtime::decoder::DecoderState::TensorPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::dtype"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState34allocateSpeculativeDecodingBuffersE23SpeculativeDecodingModeN8nvinfer18DataTypeERK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::allocateSpeculativeDecodingBuffers::speculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", "tensorrt_llm::runtime::decoder::DecoderState::disableLookahead"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState16disableLookaheadERK13RequestVector", "tensorrt_llm::runtime::decoder::DecoderState::disableLookahead::genRequests"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24getAcceptedLengthsCumSumEv", "tensorrt_llm::runtime::decoder::DecoderState::getAcceptedLengthsCumSum"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getAcceptedPackedPathsEv", "tensorrt_llm::runtime::decoder::DecoderState::getAcceptedPackedPaths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getAllNewTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getAllNewTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getBeamSearchBuffersEv", "tensorrt_llm::runtime::decoder::DecoderState::getBeamSearchBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsEv", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getCumLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getCumLogProbs::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishReasonsEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishReasons"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState16getFinishedStepsEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishedSteps"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getFinishedSumEv", "tensorrt_llm::runtime::decoder::DecoderState::getFinishedSum"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14getGatheredIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getGatheredIds::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getIds"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState6getIdsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getIds::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState21getJointDecodingInputEv", "tensorrt_llm::runtime::decoder::DecoderState::getJointDecodingInput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState22getJointDecodingOutputEv", "tensorrt_llm::runtime::decoder::DecoderState::getJointDecodingOutput"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsEv", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState11getLogProbsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getLogProbs::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBatchSizeEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxBatchSize"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState15getMaxBeamWidthEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxBeamWidth"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState27getMaxDecodingDecoderTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxDecodingDecoderTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getMaxDecodingEngineTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxDecodingEngineTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState20getMaxSequenceLengthEv", "tensorrt_llm::runtime::decoder::DecoderState::getMaxSequenceLength"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getNextDraftTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getNextDraftTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getNextDraftTokensLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getNextDraftTokensLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensEv", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getNumDecodingEngineTokensE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getNumDecodingEngineTokens::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState12getParentIdsEv", "tensorrt_llm::runtime::decoder::DecoderState::getParentIds"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState25getPrevDraftTokensLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getPrevDraftTokensLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getSequenceLengths"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsEv", "tensorrt_llm::runtime::decoder::DecoderState::getSequenceLengths"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState18getSequenceLengthsE10SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::getSequenceLengths::batchIdx"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState26getSpeculativeDecodingModeEv", "tensorrt_llm::runtime::decoder::DecoderState::getSpeculativeDecodingMode"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mBeamSearchBuffersE", "tensorrt_llm::runtime::decoder::DecoderState::mBeamSearchBuffers"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState14mFinishedStepsE", "tensorrt_llm::runtime::decoder::DecoderState::mFinishedSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState19mJointDecodingInputE", "tensorrt_llm::runtime::decoder::DecoderState::mJointDecodingInput"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState20mJointDecodingOutputE", "tensorrt_llm::runtime::decoder::DecoderState::mJointDecodingOutput"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBatchSizeE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxBatchSize"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState13mMaxBeamWidthE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxBeamWidth"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState25mMaxDecodingDecoderTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxDecodingDecoderTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mMaxDecodingEngineTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxDecodingEngineTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState18mMaxSequenceLengthE", "tensorrt_llm::runtime::decoder::DecoderState::mMaxSequenceLength"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mNumDecodingEngineTokensE", "tensorrt_llm::runtime::decoder::DecoderState::mNumDecodingEngineTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24mSpeculativeDecodingModeE", "tensorrt_llm::runtime::decoder::DecoderState::mSpeculativeDecodingMode"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens::batchIdx"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState26setNumDecodingEngineTokensE10SizeType3210SizeType32", "tensorrt_llm::runtime::decoder::DecoderState::setNumDecodingEngineTokens::numTokens"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxAttentionWindow"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxBatchSize"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxBeamWidth"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::maxSequenceLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::sinkTokenLength"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState5setupE10SizeType3210SizeType3210SizeType3210SizeType3210SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setup::worldConfig"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupEagle"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState10setupEagleEN12EagleBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupEagle::eagleBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupExplicitDraftTokens"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState24setupExplicitDraftTokensEN26ExplicitDraftTokensBuffers6InputsE", "tensorrt_llm::runtime::decoder::DecoderState::setupExplicitDraftTokens::explicitDraftTokensBuffers"], [1, 3, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", "tensorrt_llm::runtime::decoder::DecoderState::setupLookahead"], [1, 4, 1, "_CPPv4NK12tensorrt_llm7runtime7decoder12DecoderState14setupLookaheadE24LookaheadDecodingBuffers", "tensorrt_llm::runtime::decoder::DecoderState::setupLookahead::lookaheadDecodingBuffers"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::bufferManager"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::maxTokensPerEngineStep"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::modelConfig"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::speculativeDecodingMode"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime7decoder12DecoderState24setupSpeculativeDecodingERK23SpeculativeDecodingMode10SizeType32RK11ModelConfigRK11WorldConfigRK13BufferManager", "tensorrt_llm::runtime::decoder::DecoderState::setupSpeculativeDecoding::worldConfig"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", "tensorrt_llm::runtime::decoder_batch"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batchE", "tensorrt_llm::runtime::decoder_batch"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5InputE", "tensorrt_llm::runtime::decoder_batch::Input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", "tensorrt_llm::runtime::decoder_batch::Input::Input"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorI14TensorConstPtrEE", "tensorrt_llm::runtime::decoder_batch::Input::Input::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input::logits"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input5InputERKNSt6vectorINSt6vectorI14TensorConstPtrEEEE10SizeType32", "tensorrt_llm::runtime::decoder_batch::Input::Input::maxDecoderSteps"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input14TensorConstPtrE", "tensorrt_llm::runtime::decoder_batch::Input::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Input::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input10batchSlotsE", "tensorrt_llm::runtime::decoder_batch::Input::batchSlots"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input22batchSlotsRequestOrderE", "tensorrt_llm::runtime::decoder_batch::Input::batchSlotsRequestOrder"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input16cacheIndirectionE", "tensorrt_llm::runtime::decoder_batch::Input::cacheIndirection"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input11eagleInputsE", "tensorrt_llm::runtime::decoder_batch::Input::eagleInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15eagleLastInputsE", "tensorrt_llm::runtime::decoder_batch::Input::eagleLastInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input25explicitDraftTokensInputsE", "tensorrt_llm::runtime::decoder_batch::Input::explicitDraftTokensInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input29explicitDraftTokensLastInputsE", "tensorrt_llm::runtime::decoder_batch::Input::explicitDraftTokensLastInputs"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15generationStepsE", "tensorrt_llm::runtime::decoder_batch::Input::generationSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input6logitsE", "tensorrt_llm::runtime::decoder_batch::Input::logits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input15maxDecoderStepsE", "tensorrt_llm::runtime::decoder_batch::Input::maxDecoderSteps"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch5Input20predictedDraftLogitsE", "tensorrt_llm::runtime::decoder_batch::Input::predictedDraftLogits"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6OutputE", "tensorrt_llm::runtime::decoder_batch::Output"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output6OutputEv", "tensorrt_llm::runtime::decoder_batch::Output::Output"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Output::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch6Output16cacheIndirectionE", "tensorrt_llm::runtime::decoder_batch::Output::cacheIndirection"], [1, 2, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7RequestE", "tensorrt_llm::runtime::decoder_batch::Request"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9BufferPtrE", "tensorrt_llm::runtime::decoder_batch::Request::BufferPtr"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::endId"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::ids"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::inputLen"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request7RequestE14TensorConstPtr10SizeType32NSt8optionalI10SizeType32EENSt8optionalI10SizeType32EE", "tensorrt_llm::runtime::decoder_batch::Request::Request::maxNewTokens"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request14TensorConstPtrE", "tensorrt_llm::runtime::decoder_batch::Request::TensorConstPtr"], [1, 1, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request9TensorPtrE", "tensorrt_llm::runtime::decoder_batch::Request::TensorPtr"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12badWordsListE", "tensorrt_llm::runtime::decoder_batch::Request::badWordsList"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftLogitsE", "tensorrt_llm::runtime::decoder_batch::Request::draftLogits"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11draftTokensE", "tensorrt_llm::runtime::decoder_batch::Request::draftTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5dtypeE", "tensorrt_llm::runtime::decoder_batch::Request::dtype"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11eagleConfigE", "tensorrt_llm::runtime::decoder_batch::Request::eagleConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13embeddingBiasE", "tensorrt_llm::runtime::decoder_batch::Request::embeddingBias"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request5endIdE", "tensorrt_llm::runtime::decoder_batch::Request::endId"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request28generatedTokensPerEngineStepE", "tensorrt_llm::runtime::decoder_batch::Request::generatedTokensPerEngineStep"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request3idsE", "tensorrt_llm::runtime::decoder_batch::Request::ids"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request8inputLenE", "tensorrt_llm::runtime::decoder_batch::Request::inputLen"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request22lookaheadRuntimeConfigE", "tensorrt_llm::runtime::decoder_batch::Request::lookaheadRuntimeConfig"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request12maxNewTokensE", "tensorrt_llm::runtime::decoder_batch::Request::maxNewTokens"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request11medusaPathsE", "tensorrt_llm::runtime::decoder_batch::Request::medusaPaths"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13medusaTreeIdsE", "tensorrt_llm::runtime::decoder_batch::Request::medusaTreeIds"], [1, 5, 1, "_CPPv4N12tensorrt_llm7runtime13decoder_batch7Request13stopWordsListE", "tensorrt_llm::runtime::decoder_batch::Request::stopWordsList"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", "tensorrt_llm::runtime::getDefaultBatchSlots"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20getDefaultBatchSlotsEN7runtime10SizeType32E", "tensorrt_llm::runtime::getDefaultBatchSlots::batchSize"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate::ranks"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime15ipcNvlsAllocateE6size_tNSt3setIiEE", "tensorrt_llm::runtime::ipcNvlsAllocate::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", "tensorrt_llm::runtime::ipcNvlsFree"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime11ipcNvlsFreeEP13IpcNvlsHandle", "tensorrt_llm::runtime::ipcNvlsFree::handle"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime16ipcNvlsSupportedEv", "tensorrt_llm::runtime::ipcNvlsSupported"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_0"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_1"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::buffer_2"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime20lamportInitializeAllEPvPvPv6size_t", "tensorrt_llm::runtime::lamportInitializeAll::size"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<::buffer"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<::c"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<::dims"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<::module"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::operator<<::os"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<::os"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK10LoraModule", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7IBuffer", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN7ITensor5ShapeE", "tensorrt_llm::runtime::operator<<::output"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERK7ITensor", "tensorrt_llm::runtime::operator<<::tensor"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtimelsERNSt7ostreamERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::operator<<::v"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::to_string"], [1, 3, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::to_string"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERK26LoraCachePageManagerConfig", "tensorrt_llm::runtime::to_string::c"], [1, 4, 1, "_CPPv4N12tensorrt_llm7runtime9to_stringERKN9LoraCache21TaskLayerModuleConfigE", "tensorrt_llm::runtime::to_string::v"], [87, 9, 0, "-", "tensorrt_llm"]], "tensorrt_llm": [[82, 9, 0, "-", "functional"], [84, 9, 0, "-", "models"], [85, 9, 0, "-", "plugin"], [86, 9, 0, "-", "quantization"], [87, 9, 0, "-", "runtime"]], "tensorrt_llm.functional": [[82, 10, 1, "", "AllReduceFusionOp"], [82, 10, 1, "", "AllReduceParams"], [82, 10, 1, "", "AllReduceStrategy"], [82, 10, 1, "", "AttentionMaskType"], [82, 10, 1, "", "Conditional"], [82, 10, 1, "", "DimRange"], [82, 10, 1, "", "LayerNormPositionType"], [82, 10, 1, "", "LayerNormType"], [82, 10, 1, "", "MLPType"], [82, 10, 1, "", "PositionEmbeddingType"], [82, 10, 1, "", "RopeEmbeddingUtils"], [82, 10, 1, "", "RotaryScalingType"], [82, 10, 1, "", "SideStreamIDType"], [82, 10, 1, "", "SliceInputType"], [82, 10, 1, "", "Tensor"], [82, 14, 1, "", "abs"], [82, 14, 1, "", "activation"], [82, 14, 1, "", "add"], [82, 14, 1, "", "allgather"], [82, 14, 1, "", "allreduce"], [82, 14, 1, "", "arange"], [82, 14, 1, "", "argmax"], [82, 14, 1, "", "assertion"], [82, 14, 1, "", "avg_pool2d"], [82, 14, 1, "", "bert_attention"], [82, 14, 1, "", "broadcast_helper"], [82, 14, 1, "", "cast"], [82, 14, 1, "", "categorical_sample"], [82, 14, 1, "", "chunk"], [82, 14, 1, "", "clip"], [82, 14, 1, "", "concat"], [82, 14, 1, "", "constant"], [82, 14, 1, "", "constant_to_tensor_"], [82, 14, 1, "", "constants_to_tensors_"], [82, 14, 1, "", "conv1d"], [82, 14, 1, "", "conv2d"], [82, 14, 1, "", "conv3d"], [82, 14, 1, "", "conv_transpose2d"], [82, 14, 1, "", "cos"], [82, 14, 1, "", "cp_split_plugin"], [82, 14, 1, "", "create_allreduce_plugin"], [82, 14, 1, "", "cuda_stream_sync"], [82, 14, 1, "", "cumsum"], [82, 14, 1, "", "div"], [82, 14, 1, "", "dora_plugin"], [82, 14, 1, "", "einsum"], [82, 14, 1, "", "elementwise_binary"], [82, 14, 1, "", "embedding"], [82, 14, 1, "", "eq"], [82, 14, 1, "", "exp"], [82, 14, 1, "", "expand"], [82, 14, 1, "", "expand_dims"], [82, 14, 1, "", "expand_dims_like"], [82, 14, 1, "", "expand_mask"], [82, 14, 1, "", "flatten"], [82, 14, 1, "", "flip"], [82, 14, 1, "", "floordiv"], [82, 14, 1, "", "gather"], [82, 14, 1, "", "gather_last_token_logits"], [82, 14, 1, "", "gather_nd"], [82, 14, 1, "", "gegelu"], [82, 14, 1, "", "geglu"], [82, 14, 1, "", "gelu"], [82, 14, 1, "", "gemm_allreduce"], [82, 14, 1, "", "gemm_swiglu"], [82, 14, 1, "", "generate_alibi_biases"], [82, 14, 1, "", "generate_alibi_slopes"], [82, 14, 1, "", "generate_logn_scaling"], [82, 14, 1, "", "gpt_attention"], [82, 14, 1, "", "group_norm"], [82, 14, 1, "", "gt"], [82, 14, 1, "", "identity"], [82, 14, 1, "", "index_select"], [82, 14, 1, "", "int_clip"], [82, 14, 1, "", "interpolate"], [82, 14, 1, "", "is_gated_activation"], [82, 14, 1, "", "layer_norm"], [82, 14, 1, "", "log"], [82, 14, 1, "", "log_softmax"], [82, 14, 1, "", "lora_plugin"], [82, 14, 1, "", "low_latency_gemm"], [82, 14, 1, "", "low_latency_gemm_swiglu"], [82, 14, 1, "", "lt"], [82, 14, 1, "", "mamba_conv1d"], [82, 14, 1, "", "masked_scatter"], [82, 14, 1, "", "masked_select"], [82, 14, 1, "", "matmul"], [82, 14, 1, "", "max"], [82, 14, 1, "", "maximum"], [82, 14, 1, "", "mean"], [82, 14, 1, "", "meshgrid2d"], [82, 14, 1, "", "min"], [82, 14, 1, "", "minimum"], [82, 14, 1, "", "modulo"], [82, 14, 1, "", "mul"], [82, 14, 1, "", "non_gated_version"], [82, 14, 1, "", "nonzero"], [82, 14, 1, "", "not_op"], [82, 14, 1, "", "op_and"], [82, 14, 1, "", "op_or"], [82, 14, 1, "", "op_xor"], [82, 14, 1, "", "outer"], [82, 14, 1, "", "pad"], [82, 14, 1, "", "permute"], [82, 14, 1, "", "pow"], [82, 14, 1, "", "prod"], [82, 14, 1, "", "quick_gelu"], [82, 14, 1, "", "rand"], [82, 14, 1, "", "rearrange"], [82, 14, 1, "", "recv"], [82, 14, 1, "", "reduce"], [82, 14, 1, "", "reduce_scatter"], [82, 14, 1, "", "relu"], [82, 14, 1, "", "repeat"], [82, 14, 1, "", "repeat_interleave"], [82, 14, 1, "", "rg_lru"], [82, 14, 1, "", "rms_norm"], [82, 14, 1, "", "round"], [82, 14, 1, "", "scatter"], [82, 14, 1, "", "scatter_nd"], [82, 14, 1, "", "select"], [82, 14, 1, "", "selective_scan"], [82, 14, 1, "", "send"], [82, 14, 1, "", "shape"], [82, 14, 1, "", "sigmoid"], [82, 14, 1, "", "silu"], [82, 14, 1, "", "sin"], [82, 14, 1, "", "slice"], [82, 14, 1, "", "softmax"], [82, 14, 1, "", "softplus"], [82, 14, 1, "", "split"], [82, 14, 1, "", "sqrt"], [82, 14, 1, "", "squared_relu"], [82, 14, 1, "", "squeeze"], [82, 14, 1, "", "stack"], [82, 14, 1, "", "sub"], [82, 14, 1, "", "sum"], [82, 14, 1, "", "swiglu"], [82, 14, 1, "", "tanh"], [82, 14, 1, "", "topk"], [82, 14, 1, "", "transpose"], [82, 14, 1, "", "unary"], [82, 14, 1, "", "unbind"], [82, 14, 1, "", "unsqueeze"], [82, 14, 1, "", "view"], [82, 14, 1, "", "where"]], "tensorrt_llm.functional.AllReduceFusionOp": [[82, 11, 1, "", "LAST_PROCESS_FOR_UB"], [82, 11, 1, "", "MOE_ALLREDUCE_RESIDUAL_RMS_NORM"], [82, 11, 1, "", "NONE"], [82, 11, 1, "", "RESIDUAL_RMS_NORM"], [82, 11, 1, "", "RESIDUAL_RMS_NORM_OUT_QUANT_FP8"], [82, 11, 1, "", "RESIDUAL_RMS_NORM_OUT_QUANT_NVFP4"], [82, 11, 1, "", "RESIDUAL_RMS_NORM_QUANT_FP8"], [82, 11, 1, "", "RESIDUAL_RMS_NORM_QUANT_NVFP4"], [82, 11, 1, "", "RESIDUAL_RMS_PREPOST_NORM"]], "tensorrt_llm.functional.AllReduceParams": [[82, 12, 1, "", "has_affine"], [82, 12, 1, "", "has_bias"], [82, 12, 1, "", "has_scale"], [82, 12, 1, "", "update_strategy"]], "tensorrt_llm.functional.AllReduceStrategy": [[82, 11, 1, "", "AUTO"], [82, 11, 1, "", "LOWPRECISION"], [82, 11, 1, "", "MIN_LATENCY"], [82, 11, 1, "", "NCCL"], [82, 11, 1, "", "ONESHOT"], [82, 11, 1, "", "TWOSHOT"], [82, 11, 1, "", "UB"]], "tensorrt_llm.functional.AttentionMaskType": [[82, 11, 1, "", "bidirectional"], [82, 11, 1, "", "bidirectionalglm"], [82, 11, 1, "", "blocksparse"], [82, 11, 1, "", "causal"], [82, 11, 1, "", "custom_mask"], [82, 11, 1, "", "padding"], [82, 11, 1, "", "sliding_window_causal"]], "tensorrt_llm.functional.Conditional": [[82, 12, 1, "", "add_input"], [82, 12, 1, "", "add_output"]], "tensorrt_llm.functional.LayerNormPositionType": [[82, 11, 1, "", "post_layernorm"], [82, 11, 1, "", "pre_layernorm"]], "tensorrt_llm.functional.LayerNormType": [[82, 11, 1, "", "GroupNorm"], [82, 11, 1, "", "LayerNorm"], [82, 11, 1, "", "RmsNorm"]], "tensorrt_llm.functional.MLPType": [[82, 11, 1, "", "FusedGatedMLP"], [82, 11, 1, "", "GatedMLP"], [82, 11, 1, "", "MLP"]], "tensorrt_llm.functional.PositionEmbeddingType": [[82, 11, 1, "", "alibi"], [82, 11, 1, "", "alibi_with_scale"], [82, 11, 1, "", "chatglm"], [82, 12, 1, "", "choices"], [82, 11, 1, "", "deferred"], [82, 12, 1, "", "from_string"], [82, 12, 1, "", "is_alibi"], [82, 12, 1, "", "is_deferred"], [82, 12, 1, "", "is_mrope"], [82, 12, 1, "", "is_rope"], [82, 11, 1, "", "learned_absolute"], [82, 11, 1, "", "long_rope"], [82, 11, 1, "", "mrope"], [82, 11, 1, "", "relative"], [82, 11, 1, "", "rope_gpt_neox"], [82, 11, 1, "", "rope_gptj"], [82, 11, 1, "", "yarn"]], "tensorrt_llm.functional.RopeEmbeddingUtils": [[82, 12, 1, "", "apply_llama3_scaling"], [82, 12, 1, "", "apply_rotary_pos_emb"], [82, 12, 1, "", "apply_rotary_pos_emb_chatglm"], [82, 12, 1, "", "apply_rotary_pos_emb_cogvlm"], [82, 12, 1, "", "create_fake_weight"], [82, 12, 1, "", "create_sinusoidal_positions"], [82, 12, 1, "", "create_sinusoidal_positions_for_attention_plugin"], [82, 12, 1, "", "create_sinusoidal_positions_for_cogvlm_attention_plugin"], [82, 12, 1, "", "create_sinusoidal_positions_long_rope"], [82, 12, 1, "", "create_sinusoidal_positions_yarn"], [82, 12, 1, "", "rotate_every_two"], [82, 12, 1, "", "rotate_half"]], "tensorrt_llm.functional.RotaryScalingType": [[82, 11, 1, "", "dynamic"], [82, 12, 1, "", "from_string"], [82, 11, 1, "", "linear"], [82, 11, 1, "", "llama3"], [82, 11, 1, "", "longrope"], [82, 11, 1, "", "mrope"], [82, 11, 1, "", "none"], [82, 11, 1, "", "yarn"]], "tensorrt_llm.functional.SideStreamIDType": [[82, 11, 1, "", "disable"], [82, 11, 1, "", "moe"]], "tensorrt_llm.functional.SliceInputType": [[82, 11, 1, "", "axes"], [82, 11, 1, "", "data"], [82, 11, 1, "", "fill_value"], [82, 11, 1, "", "size"], [82, 11, 1, "", "start"], [82, 11, 1, "", "stride"]], "tensorrt_llm.functional.Tensor": [[82, 12, 1, "", "abs"], [82, 12, 1, "", "cast"], [82, 13, 1, "", "dtype"], [82, 12, 1, "", "flatten"], [82, 12, 1, "", "get_parent"], [82, 12, 1, "", "get_users"], [82, 12, 1, "", "is_dynamic"], [82, 12, 1, "", "is_trt_wrapper"], [82, 13, 1, "", "location"], [82, 12, 1, "", "log"], [82, 12, 1, "", "mark_output"], [82, 12, 1, "", "max"], [82, 12, 1, "", "mean"], [82, 13, 1, "", "name"], [82, 12, 1, "", "ndim"], [82, 13, 1, "", "network"], [82, 12, 1, "", "permute"], [82, 12, 1, "", "rank"], [82, 12, 1, "", "repeat"], [82, 12, 1, "", "replace_all_uses_with"], [82, 12, 1, "", "select"], [82, 13, 1, "", "shape"], [82, 12, 1, "", "size"], [82, 12, 1, "", "split"], [82, 12, 1, "", "sqrt"], [82, 12, 1, "", "squeeze"], [82, 12, 1, "", "transpose"], [82, 12, 1, "", "unbind"], [82, 12, 1, "", "unsqueeze"], [82, 12, 1, "", "view"]], "tensorrt_llm.layers": [[83, 9, 0, "-", "activation"], [83, 9, 0, "-", "attention"], [83, 9, 0, "-", "cast"], [83, 9, 0, "-", "conv"], [83, 9, 0, "-", "embedding"], [83, 9, 0, "-", "linear"], [83, 9, 0, "-", "mlp"], [83, 9, 0, "-", "normalization"], [83, 9, 0, "-", "pooling"]], "tensorrt_llm.layers.activation": [[83, 10, 1, "", "Mish"]], "tensorrt_llm.layers.activation.Mish": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention": [[83, 10, 1, "", "Attention"], [83, 10, 1, "", "AttentionMaskParams"], [83, 10, 1, "", "AttentionParams"], [83, 10, 1, "", "BertAttention"], [83, 10, 1, "", "BlockSparseAttnParams"], [83, 10, 1, "", "CogVLMAttention"], [83, 10, 1, "", "DeepseekV2Attention"], [83, 10, 1, "", "DiffusersAttention"], [83, 10, 1, "", "KeyValueCacheParams"], [83, 10, 1, "", "MropeParams"], [83, 10, 1, "", "SpecDecodingParams"], [83, 14, 1, "", "compute_relative_bias"], [83, 14, 1, "", "make_causal_mask"]], "tensorrt_llm.layers.attention.Attention": [[83, 12, 1, "", "create_attention_const_params"], [83, 12, 1, "", "fill_attention_params"], [83, 12, 1, "", "forward"], [83, 12, 1, "", "postprocess"], [83, 12, 1, "", "set_rel_attn_table"]], "tensorrt_llm.layers.attention.AttentionParams": [[83, 12, 1, "", "fill_attention_const_params_for_long_rope"], [83, 12, 1, "", "fill_attention_const_params_for_rope"], [83, 12, 1, "", "is_valid"], [83, 12, 1, "", "is_valid_cross_attn"]], "tensorrt_llm.layers.attention.BertAttention": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention.CogVLMAttention": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.attention.DeepseekV2Attention": [[83, 12, 1, "", "forward"], [83, 12, 1, "", "postprocess"], [83, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.attention.DiffusersAttention": [[83, 12, 1, "", "forward"], [83, 12, 1, "", "joint_attn_forward"]], "tensorrt_llm.layers.attention.KeyValueCacheParams": [[83, 12, 1, "", "fill_none_tensor_list"], [83, 12, 1, "", "get_first_past_key_value"], [83, 12, 1, "", "is_valid"]], "tensorrt_llm.layers.cast": [[83, 10, 1, "", "Cast"]], "tensorrt_llm.layers.cast.Cast": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv": [[83, 10, 1, "", "Conv1d"], [83, 10, 1, "", "Conv2d"], [83, 10, 1, "", "Conv3d"], [83, 10, 1, "", "ConvTranspose2d"]], "tensorrt_llm.layers.conv.Conv1d": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.Conv2d": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.Conv3d": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.conv.ConvTranspose2d": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding": [[83, 10, 1, "", "CombinedTimestepLabelEmbeddings"], [83, 10, 1, "", "CombinedTimestepTextProjEmbeddings"], [83, 10, 1, "", "Embedding"], [83, 10, 1, "", "LabelEmbedding"], [83, 10, 1, "", "PixArtAlphaTextProjection"], [83, 10, 1, "", "PromptTuningEmbedding"], [83, 10, 1, "", "SD3PatchEmbed"], [83, 10, 1, "", "TimestepEmbedding"], [83, 10, 1, "", "Timesteps"], [83, 14, 1, "", "get_1d_sincos_pos_embed_from_grid"], [83, 14, 1, "", "get_2d_sincos_pos_embed"], [83, 14, 1, "", "get_2d_sincos_pos_embed_from_grid"], [83, 14, 1, "", "get_timestep_embedding"]], "tensorrt_llm.layers.embedding.CombinedTimestepLabelEmbeddings": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.CombinedTimestepTextProjEmbeddings": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.Embedding": [[83, 12, 1, "", "forward"], [83, 12, 1, "", "postprocess"], [83, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.embedding.LabelEmbedding": [[83, 12, 1, "", "forward"], [83, 12, 1, "", "token_drop"]], "tensorrt_llm.layers.embedding.PixArtAlphaTextProjection": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.PromptTuningEmbedding": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.SD3PatchEmbed": [[83, 12, 1, "", "cropped_pos_embed"], [83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.TimestepEmbedding": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.embedding.Timesteps": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.linear": [[83, 11, 1, "", "ColumnLinear"], [83, 10, 1, "", "Linear"], [83, 10, 1, "", "LinearBase"], [83, 10, 1, "", "RowLinear"]], "tensorrt_llm.layers.linear.Linear": [[83, 12, 1, "", "collect_and_bias"], [83, 12, 1, "", "postprocess"], [83, 12, 1, "", "tp_split_dim"]], "tensorrt_llm.layers.linear.LinearBase": [[83, 12, 1, "", "collect_and_bias"], [83, 12, 1, "", "forward"], [83, 12, 1, "", "get_weight"], [83, 12, 1, "", "multiply_and_lora"], [83, 12, 1, "", "multiply_collect"], [83, 12, 1, "", "tp_split_dim"], [83, 12, 1, "", "weight_loader"]], "tensorrt_llm.layers.linear.RowLinear": [[83, 12, 1, "", "collect_and_bias"], [83, 12, 1, "", "multiply_collect"], [83, 12, 1, "", "tp_split_dim"]], "tensorrt_llm.layers.mlp": [[83, 10, 1, "", "FusedGatedMLP"], [83, 10, 1, "", "GatedMLP"], [83, 10, 1, "", "LinearActivation"], [83, 10, 1, "", "LinearApproximateGELU"], [83, 10, 1, "", "LinearGEGLU"], [83, 10, 1, "", "LinearGELU"], [83, 10, 1, "", "LinearSwiGLU"], [83, 10, 1, "", "MLP"], [83, 14, 1, "", "fc_gate_dora"], [83, 14, 1, "", "fc_gate_lora"]], "tensorrt_llm.layers.mlp.FusedGatedMLP": [[83, 12, 1, "", "fc_gate"], [83, 12, 1, "", "fc_gate_plugin"], [83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.GatedMLP": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearActivation": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearApproximateGELU": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearGEGLU": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearGELU": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.LinearSwiGLU": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.mlp.MLP": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization": [[83, 10, 1, "", "AdaLayerNorm"], [83, 10, 1, "", "AdaLayerNormContinuous"], [83, 10, 1, "", "AdaLayerNormZero"], [83, 10, 1, "", "AdaLayerNormZeroSingle"], [83, 10, 1, "", "GroupNorm"], [83, 10, 1, "", "LayerNorm"], [83, 10, 1, "", "RmsNorm"], [83, 10, 1, "", "SD35AdaLayerNormZeroX"]], "tensorrt_llm.layers.normalization.AdaLayerNorm": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormContinuous": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormZero": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.AdaLayerNormZeroSingle": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.GroupNorm": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.LayerNorm": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.RmsNorm": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.normalization.SD35AdaLayerNormZeroX": [[83, 12, 1, "", "forward"]], "tensorrt_llm.layers.pooling": [[83, 10, 1, "", "AvgPool2d"]], "tensorrt_llm.layers.pooling.AvgPool2d": [[83, 12, 1, "", "forward"]], "tensorrt_llm.llmapi": [[70, 10, 1, "", "BatchingType"], [70, 10, 1, "", "BuildCacheConfig"], [70, 10, 1, "", "BuildConfig"], [70, 10, 1, "", "CacheTransceiverConfig"], [70, 10, 1, "", "CalibConfig"], [70, 10, 1, "", "CapacitySchedulerPolicy"], [70, 10, 1, "", "CompletionOutput"], [70, 10, 1, "", "ContextChunkingPolicy"], [70, 10, 1, "", "DisaggregatedParams"], [70, 10, 1, "", "DynamicBatchConfig"], [70, 10, 1, "", "EagleDecodingConfig"], [70, 10, 1, "", "ExtendedRuntimePerfKnobConfig"], [70, 10, 1, "", "GuidedDecodingParams"], [70, 10, 1, "", "KvCacheConfig"], [70, 10, 1, "", "KvCacheRetentionConfig"], [70, 10, 1, "", "LLM"], [70, 11, 1, "", "LlmArgs"], [70, 10, 1, "", "LookaheadDecodingConfig"], [70, 10, 1, "", "MTPDecodingConfig"], [70, 10, 1, "", "MedusaDecodingConfig"], [70, 10, 1, "", "MpiCommSession"], [70, 10, 1, "", "NGramDecodingConfig"], [70, 10, 1, "", "QuantAlgo"], [70, 10, 1, "", "QuantConfig"], [70, 10, 1, "", "RequestError"], [70, 10, 1, "", "RequestOutput"], [70, 10, 1, "", "SamplingParams"], [70, 10, 1, "", "SchedulerConfig"], [70, 10, 1, "", "TorchLlmArgs"], [70, 10, 1, "", "TrtLlmArgs"]], "tensorrt_llm.llmapi.BatchingType": [[70, 11, 1, "", "INFLIGHT"], [70, 11, 1, "", "STATIC"]], "tensorrt_llm.llmapi.BuildCacheConfig": [[70, 12, 1, "", "__init__"], [70, 13, 1, "id7", "cache_root"], [70, 13, 1, "id8", "max_cache_storage_gb"], [70, 13, 1, "id9", "max_records"]], "tensorrt_llm.llmapi.BuildConfig": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "auto_parallel_config"], [70, 11, 1, "", "dry_run"], [70, 11, 1, "", "enable_debug_output"], [70, 11, 1, "", "force_num_profiles"], [70, 12, 1, "", "from_dict"], [70, 12, 1, "", "from_json_file"], [70, 11, 1, "", "gather_context_logits"], [70, 11, 1, "", "gather_generation_logits"], [70, 11, 1, "", "input_timing_cache"], [70, 11, 1, "", "kv_cache_type"], [70, 11, 1, "", "lora_config"], [70, 11, 1, "", "max_batch_size"], [70, 11, 1, "", "max_beam_width"], [70, 11, 1, "", "max_draft_len"], [70, 11, 1, "", "max_encoder_input_len"], [70, 11, 1, "", "max_input_len"], [70, 11, 1, "", "max_num_tokens"], [70, 11, 1, "", "max_prompt_embedding_table_size"], [70, 11, 1, "", "max_seq_len"], [70, 11, 1, "", "monitor_memory"], [70, 11, 1, "", "opt_batch_size"], [70, 11, 1, "", "opt_num_tokens"], [70, 11, 1, "", "output_timing_cache"], [70, 11, 1, "", "plugin_config"], [70, 11, 1, "", "profiling_verbosity"], [70, 11, 1, "", "speculative_decoding_mode"], [70, 11, 1, "", "strongly_typed"], [70, 12, 1, "", "to_dict"], [70, 12, 1, "", "update"], [70, 12, 1, "", "update_from_dict"], [70, 12, 1, "", "update_kv_cache_type"], [70, 11, 1, "", "use_mrope"], [70, 11, 1, "", "use_refit"], [70, 11, 1, "", "use_strip_plan"], [70, 11, 1, "", "visualize_network"], [70, 11, 1, "", "weight_sparsity"], [70, 11, 1, "", "weight_streaming"]], "tensorrt_llm.llmapi.CacheTransceiverConfig": [[70, 15, 1, "", "max_num_tokens"], [70, 11, 1, "", "model_config"]], "tensorrt_llm.llmapi.CalibConfig": [[70, 15, 1, "", "calib_batch_size"], [70, 15, 1, "", "calib_batches"], [70, 15, 1, "", "calib_dataset"], [70, 15, 1, "", "calib_max_seq_length"], [70, 15, 1, "", "device"], [70, 12, 1, "", "from_dict"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "random_seed"], [70, 12, 1, "", "to_dict"], [70, 15, 1, "", "tokenizer_max_seq_length"]], "tensorrt_llm.llmapi.CapacitySchedulerPolicy": [[70, 11, 1, "", "GUARANTEED_NO_EVICT"], [70, 11, 1, "", "MAX_UTILIZATION"], [70, 11, 1, "", "STATIC_BATCH"]], "tensorrt_llm.llmapi.CompletionOutput": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "cumulative_logprob"], [70, 11, 1, "", "disaggregated_params"], [70, 11, 1, "", "finish_reason"], [70, 11, 1, "", "generation_logits"], [70, 11, 1, "", "index"], [70, 13, 1, "id2", "length"], [70, 11, 1, "", "logprobs"], [70, 13, 1, "id3", "logprobs_diff"], [70, 11, 1, "", "prompt_logprobs"], [70, 11, 1, "", "stop_reason"], [70, 11, 1, "", "text"], [70, 13, 1, "id4", "text_diff"], [70, 11, 1, "", "token_ids"], [70, 13, 1, "id5", "token_ids_diff"]], "tensorrt_llm.llmapi.ContextChunkingPolicy": [[70, 11, 1, "", "EQUAL_PROGRESS"], [70, 11, 1, "", "FIRST_COME_FIRST_SERVED"]], "tensorrt_llm.llmapi.DisaggregatedParams": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "ctx_request_id"], [70, 11, 1, "", "draft_tokens"], [70, 11, 1, "", "first_gen_tokens"], [70, 12, 1, "", "get_context_phase_params"], [70, 12, 1, "", "get_request_type"], [70, 11, 1, "", "opaque_state"], [70, 11, 1, "", "request_type"]], "tensorrt_llm.llmapi.DynamicBatchConfig": [[70, 15, 1, "", "dynamic_batch_moving_average_window"], [70, 15, 1, "", "enable_batch_size_tuning"], [70, 15, 1, "", "enable_max_num_tokens_tuning"], [70, 11, 1, "", "model_config"]], "tensorrt_llm.llmapi.EagleDecodingConfig": [[70, 11, 1, "", "decoding_type"], [70, 15, 1, "", "dynamic_tree_max_topK"], [70, 15, 1, "", "eagle3_one_model"], [70, 15, 1, "", "eagle_choices"], [70, 12, 1, "", "from_dict"], [70, 15, 1, "", "greedy_sampling"], [70, 15, 1, "", "max_non_leaves_per_layer"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "num_eagle_layers"], [70, 15, 1, "", "posterior_threshold"], [70, 15, 1, "", "pytorch_eagle_weights_path"], [70, 15, 1, "", "use_dynamic_tree"]], "tensorrt_llm.llmapi.ExtendedRuntimePerfKnobConfig": [[70, 15, 1, "", "cuda_graph_cache_size"], [70, 15, 1, "", "cuda_graph_mode"], [70, 15, 1, "", "enable_context_fmha_fp32_acc"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "multi_block_mode"]], "tensorrt_llm.llmapi.GuidedDecodingParams": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "grammar"], [70, 11, 1, "", "json"], [70, 11, 1, "", "json_object"], [70, 11, 1, "", "regex"], [70, 11, 1, "", "structural_tag"]], "tensorrt_llm.llmapi.KvCacheConfig": [[70, 15, 1, "", "copy_on_partial_reuse"], [70, 15, 1, "", "cross_kv_cache_fraction"], [70, 15, 1, "", "enable_block_reuse"], [70, 15, 1, "", "enable_partial_reuse"], [70, 15, 1, "", "event_buffer_max_size"], [70, 15, 1, "", "free_gpu_memory_fraction"], [70, 15, 1, "", "host_cache_size"], [70, 15, 1, "", "max_attention_window"], [70, 15, 1, "", "max_tokens"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "onboard_blocks"], [70, 15, 1, "", "secondary_offload_min_priority"], [70, 15, 1, "", "sink_token_length"]], "tensorrt_llm.llmapi.KvCacheRetentionConfig": [[70, 10, 1, "", "TokenRangeRetentionConfig"], [70, 12, 1, "", "__init__"], [70, 13, 1, "", "decode_duration_ms"], [70, 13, 1, "", "decode_retention_priority"], [70, 13, 1, "", "directory"], [70, 13, 1, "", "token_range_retention_configs"], [70, 13, 1, "", "transfer_mode"]], "tensorrt_llm.llmapi.KvCacheRetentionConfig.TokenRangeRetentionConfig": [[70, 12, 1, "", "__init__"], [70, 13, 1, "", "duration_ms"], [70, 13, 1, "", "priority"], [70, 13, 1, "", "token_end"], [70, 13, 1, "", "token_start"]], "tensorrt_llm.llmapi.LLM": [[70, 12, 1, "", "__init__"], [70, 12, 1, "", "generate"], [70, 12, 1, "", "generate_async"], [70, 12, 1, "", "get_kv_cache_events"], [70, 12, 1, "", "get_kv_cache_events_async"], [70, 12, 1, "", "get_stats"], [70, 12, 1, "", "get_stats_async"], [70, 12, 1, "", "save"], [70, 12, 1, "", "shutdown"], [70, 13, 1, "id0", "tokenizer"], [70, 13, 1, "id1", "workspace"]], "tensorrt_llm.llmapi.LookaheadDecodingConfig": [[70, 12, 1, "", "__init__"], [70, 12, 1, "", "calculate_speculative_resource"], [70, 11, 1, "", "decoding_type"], [70, 12, 1, "", "from_dict"], [70, 15, 1, "", "max_ngram_size"], [70, 15, 1, "", "max_verification_set_size"], [70, 15, 1, "", "max_window_size"], [70, 11, 1, "", "model_config"], [70, 16, 1, "", "validate_positive_values"]], "tensorrt_llm.llmapi.MTPDecodingConfig": [[70, 11, 1, "", "decoding_type"], [70, 12, 1, "", "from_dict"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "num_nextn_predict_layers"], [70, 15, 1, "", "relaxed_delta"], [70, 15, 1, "", "relaxed_topk"], [70, 15, 1, "", "use_relaxed_acceptance_for_thinking"]], "tensorrt_llm.llmapi.MedusaDecodingConfig": [[70, 11, 1, "", "decoding_type"], [70, 12, 1, "", "from_dict"], [70, 15, 1, "", "medusa_choices"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "num_medusa_heads"]], "tensorrt_llm.llmapi.MpiCommSession": [[70, 12, 1, "", "__init__"], [70, 12, 1, "", "abort"], [70, 12, 1, "", "get_comm"], [70, 12, 1, "", "shutdown"], [70, 12, 1, "", "submit"], [70, 12, 1, "", "submit_sync"]], "tensorrt_llm.llmapi.NGramDecodingConfig": [[70, 11, 1, "", "decoding_type"], [70, 12, 1, "", "from_dict"], [70, 15, 1, "", "is_keep_all"], [70, 15, 1, "", "is_public_pool"], [70, 15, 1, "", "is_use_oldest"], [70, 15, 1, "", "max_matching_ngram_size"], [70, 11, 1, "", "model_config"], [70, 15, 1, "", "prompt_lookup_num_tokens"]], "tensorrt_llm.llmapi.QuantAlgo": [[70, 11, 1, "", "FP8"], [70, 11, 1, "", "FP8_BLOCK_SCALES"], [70, 11, 1, "", "FP8_PER_CHANNEL_PER_TOKEN"], [70, 11, 1, "", "INT8"], [70, 11, 1, "", "MIXED_PRECISION"], [70, 11, 1, "", "NO_QUANT"], [70, 11, 1, "", "NVFP4"], [70, 11, 1, "", "W4A16"], [70, 11, 1, "", "W4A16_AWQ"], [70, 11, 1, "", "W4A16_GPTQ"], [70, 11, 1, "", "W4A8_AWQ"], [70, 11, 1, "", "W4A8_QSERVE_PER_CHANNEL"], [70, 11, 1, "", "W4A8_QSERVE_PER_GROUP"], [70, 11, 1, "", "W8A16"], [70, 11, 1, "", "W8A16_GPTQ"], [70, 11, 1, "", "W8A8_SQ_PER_CHANNEL"], [70, 11, 1, "", "W8A8_SQ_PER_CHANNEL_PER_TENSOR_PLUGIN"], [70, 11, 1, "", "W8A8_SQ_PER_CHANNEL_PER_TOKEN_PLUGIN"], [70, 11, 1, "", "W8A8_SQ_PER_TENSOR_PER_TOKEN_PLUGIN"], [70, 11, 1, "", "W8A8_SQ_PER_TENSOR_PLUGIN"]], "tensorrt_llm.llmapi.QuantConfig": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "clamp_val"], [70, 11, 1, "", "exclude_modules"], [70, 12, 1, "", "from_dict"], [70, 11, 1, "", "group_size"], [70, 11, 1, "", "has_zero_point"], [70, 12, 1, "", "is_module_excluded_from_quantization"], [70, 11, 1, "", "kv_cache_quant_algo"], [70, 13, 1, "", "layer_quant_mode"], [70, 11, 1, "", "pre_quant_scale"], [70, 11, 1, "", "quant_algo"], [70, 13, 1, "", "quant_mode"], [70, 11, 1, "", "smoothquant_val"], [70, 12, 1, "", "to_dict"], [70, 11, 1, "", "use_meta_recipe"]], "tensorrt_llm.llmapi.RequestOutput": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "context_logits"], [70, 11, 1, "", "finished"], [70, 11, 1, "", "outputs"], [70, 13, 1, "id6", "prompt"], [70, 11, 1, "", "prompt_token_ids"], [70, 11, 1, "", "request_id"]], "tensorrt_llm.llmapi.SamplingParams": [[70, 12, 1, "", "__init__"], [70, 11, 1, "", "add_special_tokens"], [70, 11, 1, "", "additional_model_outputs"], [70, 11, 1, "", "apply_batched_logits_processor"], [70, 11, 1, "", "bad"], [70, 11, 1, "", "bad_token_ids"], [70, 11, 1, "", "beam_search_diversity_rate"], [70, 11, 1, "", "beam_width_array"], [70, 11, 1, "", "best_of"], [70, 11, 1, "", "detokenize"], [70, 11, 1, "", "early_stopping"], [70, 11, 1, "", "embedding_bias"], [70, 11, 1, "", "end_id"], [70, 11, 1, "", "exclude_input_from_output"], [70, 11, 1, "", "frequency_penalty"], [70, 11, 1, "", "guided_decoding"], [70, 11, 1, "", "ignore_eos"], [70, 11, 1, "", "include_stop_str_in_output"], [70, 11, 1, "", "length_penalty"], [70, 11, 1, "", "logits_processor"], [70, 11, 1, "", "logprobs"], [70, 11, 1, "", "lookahead_config"], [70, 11, 1, "", "max_tokens"], [70, 11, 1, "", "min_p"], [70, 11, 1, "", "min_tokens"], [70, 11, 1, "", "n"], [70, 11, 1, "", "no_repeat_ngram_size"], [70, 11, 1, "", "pad_id"], [70, 11, 1, "", "presence_penalty"], [70, 11, 1, "", "prompt_logprobs"], [70, 11, 1, "", "repetition_penalty"], [70, 11, 1, "", "return_context_logits"], [70, 11, 1, "", "return_encoder_output"], [70, 11, 1, "", "return_generation_logits"], [70, 11, 1, "", "return_perf_metrics"], [70, 11, 1, "", "seed"], [70, 11, 1, "", "skip_special_tokens"], [70, 11, 1, "", "spaces_between_special_tokens"], [70, 11, 1, "", "stop"], [70, 11, 1, "", "stop_token_ids"], [70, 11, 1, "", "temperature"], [70, 11, 1, "", "top_k"], [70, 11, 1, "", "top_p"], [70, 11, 1, "", "top_p_decay"], [70, 11, 1, "", "top_p_min"], [70, 11, 1, "", "top_p_reset_ids"], [70, 11, 1, "", "truncate_prompt_tokens"], [70, 11, 1, "", "use_beam_search"]], "tensorrt_llm.llmapi.SchedulerConfig": [[70, 15, 1, "", "capacity_scheduler_policy"], [70, 15, 1, "", "context_chunking_policy"], [70, 15, 1, "", "dynamic_batch_config"], [70, 11, 1, "", "model_config"]], "tensorrt_llm.llmapi.TorchLlmArgs": [[70, 15, 1, "", "attn_backend"], [70, 15, 1, "", "auto_deploy_config"], [70, 15, 1, "", "autotuner_enabled"], [70, 15, 1, "", "build_config"], [70, 16, 1, "", "convert_load_format"], [70, 15, 1, "", "cuda_graph_batch_sizes"], [70, 15, 1, "", "cuda_graph_max_batch_size"], [70, 15, 1, "", "cuda_graph_padding_enabled"], [70, 11, 1, "", "decoding_config"], [70, 15, 1, "", "disable_overlap_scheduler"], [70, 15, 1, "", "enable_iter_perf_stats"], [70, 15, 1, "", "enable_iter_req_stats"], [70, 15, 1, "", "enable_layerwise_nvtx_marker"], [70, 15, 1, "", "enable_min_latency"], [70, 15, 1, "", "enable_trtllm_sampler"], [70, 13, 1, "", "extra_resource_managers"], [70, 11, 1, "id18", "field_name"], [70, 12, 1, "", "get_pytorch_backend_config"], [70, 15, 1, "", "kv_cache_dtype"], [70, 15, 1, "", "load_format"], [70, 11, 1, "", "max_cpu_loras"], [70, 11, 1, "", "max_lora_rank"], [70, 11, 1, "", "max_loras"], [70, 15, 1, "", "mixed_sampler"], [70, 11, 1, "", "model_config"], [70, 12, 1, "", "model_post_init"], [70, 15, 1, "", "moe_backend"], [70, 15, 1, "", "moe_load_balancer"], [70, 15, 1, "", "moe_max_num_tokens"], [70, 11, 1, "id16", "msg"], [70, 15, 1, "", "print_iter_log"], [70, 15, 1, "", "torch_compile_enable_userbuffers"], [70, 15, 1, "", "torch_compile_enabled"], [70, 15, 1, "", "torch_compile_fullgraph"], [70, 15, 1, "", "torch_compile_inductor_enabled"], [70, 15, 1, "", "torch_compile_piecewise_cuda_graph"], [70, 15, 1, "", "use_cuda_graph"], [70, 15, 1, "", "use_kv_cache"], [70, 16, 1, "", "validate_cuda_graph_config"], [70, 16, 1, "", "validate_cuda_graph_max_batch_size"], [70, 11, 1, "id17", "wrapped_property"]], "tensorrt_llm.llmapi.TrtLlmArgs": [[70, 11, 1, "", "auto_parallel"], [70, 13, 1, "", "auto_parallel_config"], [70, 11, 1, "", "auto_parallel_world_size"], [70, 15, 1, "", "build_config"], [70, 15, 1, "", "calib_config"], [70, 11, 1, "", "decoding_config"], [70, 15, 1, "", "embedding_parallel_mode"], [70, 15, 1, "", "enable_build_cache"], [70, 15, 1, "", "enable_tqdm"], [70, 15, 1, "", "extended_runtime_perf_knob_config"], [70, 15, 1, "", "fast_build"], [70, 11, 1, "id33", "field_name"], [70, 11, 1, "", "max_cpu_loras"], [70, 11, 1, "", "max_lora_rank"], [70, 11, 1, "", "max_loras"], [70, 11, 1, "", "model_config"], [70, 12, 1, "", "model_post_init"], [70, 11, 1, "id31", "msg"], [70, 15, 1, "", "workspace"], [70, 11, 1, "id32", "wrapped_property"]], "tensorrt_llm.models": [[84, 10, 1, "", "BaichuanForCausalLM"], [84, 10, 1, "", "BertForQuestionAnswering"], [84, 10, 1, "", "BertForSequenceClassification"], [84, 10, 1, "", "BertModel"], [84, 10, 1, "", "BloomForCausalLM"], [84, 10, 1, "", "BloomModel"], [84, 10, 1, "", "CLIPVisionTransformer"], [84, 10, 1, "", "ChatGLMConfig"], [84, 10, 1, "", "ChatGLMForCausalLM"], [84, 10, 1, "", "ChatGLMModel"], [84, 10, 1, "", "CogVLMConfig"], [84, 10, 1, "", "CogVLMForCausalLM"], [84, 10, 1, "", "CohereForCausalLM"], [84, 10, 1, "", "DbrxConfig"], [84, 10, 1, "", "DbrxForCausalLM"], [84, 10, 1, "", "DecoderModel"], [84, 10, 1, "", "DeepseekForCausalLM"], [84, 10, 1, "", "DeepseekV2ForCausalLM"], [84, 10, 1, "", "DiT"], [84, 10, 1, "", "EagleForCausalLM"], [84, 10, 1, "", "EncoderModel"], [84, 10, 1, "", "FalconConfig"], [84, 10, 1, "", "FalconForCausalLM"], [84, 10, 1, "", "FalconModel"], [84, 10, 1, "", "GPTConfig"], [84, 10, 1, "", "GPTForCausalLM"], [84, 10, 1, "", "GPTJConfig"], [84, 10, 1, "", "GPTJForCausalLM"], [84, 10, 1, "", "GPTJModel"], [84, 10, 1, "", "GPTModel"], [84, 10, 1, "", "GPTNeoXForCausalLM"], [84, 10, 1, "", "GPTNeoXModel"], [84, 10, 1, "", "GemmaConfig"], [84, 10, 1, "", "GemmaForCausalLM"], [84, 10, 1, "", "LLaMAConfig"], [84, 10, 1, "", "LLaMAForCausalLM"], [84, 10, 1, "", "LLaMAModel"], [84, 10, 1, "", "LlavaNextVisionConfig"], [84, 10, 1, "", "LlavaNextVisionWrapper"], [84, 10, 1, "", "MLLaMAForCausalLM"], [84, 10, 1, "", "MPTForCausalLM"], [84, 10, 1, "", "MPTModel"], [84, 10, 1, "", "MambaForCausalLM"], [84, 10, 1, "", "MedusaConfig"], [84, 10, 1, "", "MedusaForCausalLm"], [84, 10, 1, "", "OPTForCausalLM"], [84, 10, 1, "", "OPTModel"], [84, 10, 1, "", "Phi3ForCausalLM"], [84, 10, 1, "", "Phi3Model"], [84, 10, 1, "", "PhiForCausalLM"], [84, 10, 1, "", "PhiModel"], [84, 10, 1, "", "PretrainedConfig"], [84, 10, 1, "", "PretrainedModel"], [84, 10, 1, "", "ReDrafterForCausalLM"], [84, 10, 1, "", "RecurrentGemmaForCausalLM"], [84, 11, 1, "", "RobertaForQuestionAnswering"], [84, 11, 1, "", "RobertaForSequenceClassification"], [84, 11, 1, "", "RobertaModel"], [84, 10, 1, "", "SD3Transformer2DModel"], [84, 10, 1, "", "SpeculativeDecodingMode"], [84, 10, 1, "", "WhisperEncoder"]], "tensorrt_llm.models.BaichuanForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "quantize"]], "tensorrt_llm.models.BertForQuestionAnswering": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.BertForSequenceClassification": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.BertModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.BloomModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.CLIPVisionTransformer": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.ChatGLMConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.ChatGLMForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "quantize"]], "tensorrt_llm.models.ChatGLMModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.CogVLMConfig": [[84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.CogVLMForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "default_plugin_config"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "quantize"]], "tensorrt_llm.models.CohereForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DbrxConfig": [[84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.DbrxForCausalLM": [[84, 11, 1, "", "config_class"]], "tensorrt_llm.models.DecoderModel": [[84, 12, 1, "", "check_config"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "precompute_relative_attention_bias"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.DeepseekForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DeepseekV2ForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.DiT": [[84, 12, 1, "", "check_config"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "forward_with_cfg"], [84, 12, 1, "", "forward_without_cfg"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "unpatchify"]], "tensorrt_llm.models.EagleForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.EncoderModel": [[84, 12, 1, "", "check_config"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "precompute_relative_attention_bias"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "use_lora"], [84, 12, 1, "", "use_prompt_tuning"]], "tensorrt_llm.models.FalconConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.FalconForCausalLM": [[84, 12, 1, "", "check_config"], [84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.FalconModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "from_nemo"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GPTForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "from_nemo"], [84, 12, 1, "", "quantize"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.GPTJConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GPTJForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.GPTJModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.GPTNeoXModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.GemmaConfig": [[84, 11, 1, "", "GEMMA2_ADDED_FIELDS"], [84, 11, 1, "", "GEMMA3_ADDED_FIELDS"], [84, 11, 1, "", "GEMMA_ADDED_FIELDS"], [84, 11, 1, "", "VERBATIM"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "gemma2_config"], [84, 12, 1, "", "gemma3_config"], [84, 12, 1, "", "get_hf_config"], [84, 13, 1, "", "is_gemma_2"], [84, 13, 1, "", "is_gemma_3"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.GemmaForCausalLM": [[84, 11, 1, "", "NATIVE_QUANT_FLOW"], [84, 12, 1, "", "assert_valid_quant_algo"], [84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "quantize"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.LLaMAConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "from_meta_ckpt"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.LLaMAForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "default_plugin_config"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "from_meta_ckpt"], [84, 12, 1, "", "quantize"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.LLaMAModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.LlavaNextVisionConfig": [[84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.LlavaNextVisionWrapper": [[84, 12, 1, "", "forward"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "save_checkpoint"]], "tensorrt_llm.models.MLLaMAForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.MPTForCausalLM": [[84, 12, 1, "", "check_config"]], "tensorrt_llm.models.MPTModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.MambaForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.MedusaConfig": [[84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "to_dict"]], "tensorrt_llm.models.MedusaForCausalLm": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"]], "tensorrt_llm.models.OPTForCausalLM": [[84, 12, 1, "", "check_config"]], "tensorrt_llm.models.OPTModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.Phi3ForCausalLM": [[84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.Phi3Model": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.PhiForCausalLM": [[84, 12, 1, "", "check_config"], [84, 11, 1, "", "config_class"], [84, 12, 1, "", "from_hugging_face"], [84, 12, 1, "", "use_lora"]], "tensorrt_llm.models.PhiModel": [[84, 12, 1, "", "forward"]], "tensorrt_llm.models.PretrainedConfig": [[84, 12, 1, "", "create_runtime_defaults"], [84, 12, 1, "", "for_each_rank"], [84, 12, 1, "", "from_checkpoint"], [84, 12, 1, "", "from_dict"], [84, 12, 1, "", "from_json_file"], [84, 12, 1, "", "get_config_group"], [84, 12, 1, "", "has_config_group"], [84, 13, 1, "", "kv_dtype"], [84, 13, 1, "", "quant_algo"], [84, 13, 1, "", "quant_mode"], [84, 12, 1, "", "set_if_not_exist"], [84, 12, 1, "", "set_rank"], [84, 12, 1, "", "to_dict"], [84, 12, 1, "", "to_json_file"], [84, 12, 1, "", "to_layer_quant_config"]], "tensorrt_llm.models.PretrainedModel": [[84, 12, 1, "", "check_config"], [84, 12, 1, "", "from_checkpoint"], [84, 12, 1, "", "from_config"], [84, 12, 1, "", "load"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "quantize"], [84, 12, 1, "", "release"], [84, 12, 1, "", "save_checkpoint"]], "tensorrt_llm.models.ReDrafterForCausalLM": [[84, 12, 1, "", "forward"], [84, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.models.RecurrentGemmaForCausalLM": [[84, 12, 1, "", "forward"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "prepare_recurrent_inputs"]], "tensorrt_llm.models.SD3Transformer2DModel": [[84, 13, 1, "", "attn_processors"], [84, 11, 1, "", "config_class"], [84, 12, 1, "", "disable_forward_chunking"], [84, 12, 1, "", "enable_forward_chunking"], [84, 12, 1, "", "forward"], [84, 12, 1, "", "from_pretrained"], [84, 12, 1, "", "fuse_qkv_projections"], [84, 12, 1, "", "load"], [84, 12, 1, "", "prepare_inputs"], [84, 12, 1, "", "set_attn_processor"], [84, 12, 1, "", "unfuse_qkv_projections"]], "tensorrt_llm.models.SpeculativeDecodingMode": [[84, 11, 1, "", "DRAFT_TOKENS_EXTERNAL"], [84, 11, 1, "", "EAGLE"], [84, 11, 1, "", "EXPLICIT_DRAFT_TOKENS"], [84, 11, 1, "", "LOOKAHEAD_DECODING"], [84, 11, 1, "", "MEDUSA"], [84, 11, 1, "", "NGRAM"], [84, 11, 1, "", "NONE"], [84, 12, 1, "", "from_arguments"]], "tensorrt_llm.models.WhisperEncoder": [[84, 12, 1, "", "forward"], [84, 12, 1, "", "precompute_relative_attention_bias"], [84, 12, 1, "", "prepare_inputs"]], "tensorrt_llm.plugin": [[85, 10, 1, "", "PluginConfig"]], "tensorrt_llm.plugin.PluginConfig": [[85, 12, 1, "", "to_legacy_setting"]], "tensorrt_llm.quantization": [[86, 10, 1, "", "QuantAlgo"], [86, 10, 1, "", "QuantMode"], [86, 14, 1, "", "quantize_and_export"]], "tensorrt_llm.runtime": [[87, 10, 1, "", "ChatGLMGenerationSession"], [87, 10, 1, "", "EncDecModelRunner"], [87, 10, 1, "", "GenerationSequence"], [87, 10, 1, "", "GenerationSession"], [87, 10, 1, "", "KVCacheManager"], [87, 10, 1, "", "LogitsProcessor"], [87, 10, 1, "", "LogitsProcessorList"], [87, 10, 1, "", "ModelConfig"], [87, 10, 1, "", "ModelRunner"], [87, 10, 1, "", "ModelRunnerCpp"], [87, 10, 1, "", "MultimodalModelRunner"], [87, 10, 1, "", "QWenForCausalLMGenerationSession"], [87, 10, 1, "", "SamplingConfig"], [87, 10, 1, "", "Session"], [87, 10, 1, "", "StoppingCriteria"], [87, 10, 1, "", "StoppingCriteriaList"], [87, 10, 1, "", "TensorInfo"], [87, 14, 1, "", "decode_words_list"]], "tensorrt_llm.runtime.EncDecModelRunner": [[87, 12, 1, "", "encoder_run"], [87, 12, 1, "", "from_engine"], [87, 12, 1, "", "generate"], [87, 12, 1, "", "process_input"]], "tensorrt_llm.runtime.GenerationSequence": [[87, 12, 1, "", "get_batch_idx"], [87, 12, 1, "", "get_seq_idx"]], "tensorrt_llm.runtime.GenerationSession": [[87, 11, 1, "", "batch_size"], [87, 11, 1, "", "buffer_allocated"], [87, 13, 1, "", "context_mem_size"], [87, 13, 1, "", "conv_kernel"], [87, 13, 1, "", "cross_attention"], [87, 11, 1, "", "cuda_graph_mode"], [87, 12, 1, "", "cuda_stream_guard"], [87, 11, 1, "", "debug_mode"], [87, 11, 1, "", "debug_tensors_to_save"], [87, 12, 1, "", "decode"], [87, 12, 1, "", "decode_batch"], [87, 12, 1, "", "decode_regular"], [87, 12, 1, "", "decode_stream"], [87, 11, 1, "", "device"], [87, 13, 1, "", "dtype"], [87, 12, 1, "", "dump_debug_buffers"], [87, 12, 1, "", "early_stop_criteria"], [87, 13, 1, "", "engine_inspector"], [87, 12, 1, "", "filter_medusa_logits"], [87, 12, 1, "", "finalize_decoder"], [87, 12, 1, "", "find_best_medusa_path"], [87, 13, 1, "", "first_layer"], [87, 13, 1, "", "gather_context_logits"], [87, 13, 1, "", "gather_generation_logits"], [87, 13, 1, "", "gemm_allreduce_plugin"], [87, 12, 1, "", "get_next_medusa_tokens"], [87, 12, 1, "", "get_num_heads_kv"], [87, 12, 1, "", "handle_per_step"], [87, 13, 1, "", "has_position_embedding"], [87, 13, 1, "", "has_token_type_embedding"], [87, 13, 1, "", "head_size"], [87, 13, 1, "", "hidden_size"], [87, 13, 1, "", "is_medusa_mode"], [87, 13, 1, "", "is_redrafter_mode"], [87, 13, 1, "", "kv_cache_type"], [87, 13, 1, "", "last_layer"], [87, 12, 1, "", "locate_accepted_draft_tokens"], [87, 11, 1, "", "mapping"], [87, 13, 1, "", "max_draft_tokens"], [87, 13, 1, "", "max_prompt_embedding_table_size"], [87, 12, 1, "", "medusa_decode_and_verify"], [87, 11, 1, "", "medusa_paths"], [87, 11, 1, "", "medusa_position_offsets"], [87, 11, 1, "", "medusa_temperature"], [87, 11, 1, "", "medusa_topks"], [87, 11, 1, "", "medusa_tree_ids"], [87, 12, 1, "", "next_medusa_input_ids"], [87, 11, 1, "", "num_draft_tokens"], [87, 13, 1, "", "num_heads"], [87, 13, 1, "", "num_layers"], [87, 13, 1, "", "num_medusa_heads"], [87, 13, 1, "", "paged_kv_cache"], [87, 13, 1, "", "paged_state"], [87, 12, 1, "", "pp_communicate_final_output_ids"], [87, 12, 1, "", "pp_communicate_new_tokens"], [87, 12, 1, "", "process_logits_including_draft"], [87, 13, 1, "", "profiler"], [87, 13, 1, "", "quant_mode"], [87, 13, 1, "", "remove_input_padding"], [87, 12, 1, "", "reorder_kv_cache_for_beam_search"], [87, 13, 1, "", "rnn_conv_dim_size"], [87, 13, 1, "", "rnn_head_size"], [87, 13, 1, "", "rnn_hidden_size"], [87, 11, 1, "", "runtime"], [87, 12, 1, "", "setup"], [87, 13, 1, "", "state_dtype"], [87, 13, 1, "", "state_size"], [87, 13, 1, "", "tokens_per_block"], [87, 12, 1, "", "update_output_ids_by_offset"], [87, 13, 1, "", "use_gemm_allreduce_plugin"], [87, 13, 1, "", "use_gpt_attention_plugin"], [87, 13, 1, "", "use_kv_cache"], [87, 13, 1, "", "use_lora_plugin"], [87, 13, 1, "", "use_mamba_conv1d_plugin"], [87, 13, 1, "", "vocab_size"]], "tensorrt_llm.runtime.KVCacheManager": [[87, 12, 1, "", "add_sequence"], [87, 12, 1, "", "get_block_offsets"], [87, 12, 1, "", "step"]], "tensorrt_llm.runtime.ModelConfig": [[87, 11, 1, "", "conv_kernel"], [87, 11, 1, "", "cross_attention"], [87, 11, 1, "", "dtype"], [87, 11, 1, "", "gather_context_logits"], [87, 11, 1, "", "gather_generation_logits"], [87, 11, 1, "", "gemm_allreduce_plugin"], [87, 11, 1, "", "gpt_attention_plugin"], [87, 11, 1, "", "gpu_weights_percent"], [87, 11, 1, "", "has_position_embedding"], [87, 11, 1, "", "has_token_type_embedding"], [87, 11, 1, "", "head_size"], [87, 11, 1, "", "hidden_size"], [87, 11, 1, "", "kv_cache_type"], [87, 11, 1, "", "language_adapter_config"], [87, 11, 1, "", "layer_types"], [87, 11, 1, "", "lora_plugin"], [87, 11, 1, "", "lora_target_modules"], [87, 11, 1, "", "mamba_conv1d_plugin"], [87, 11, 1, "", "max_batch_size"], [87, 11, 1, "", "max_beam_width"], [87, 11, 1, "", "max_medusa_tokens"], [87, 11, 1, "", "max_prompt_embedding_table_size"], [87, 11, 1, "", "model_name"], [87, 11, 1, "", "num_heads"], [87, 11, 1, "", "num_kv_heads"], [87, 11, 1, "", "num_kv_heads_per_cross_attn_layer"], [87, 11, 1, "", "num_kv_heads_per_layer"], [87, 11, 1, "", "num_layers"], [87, 11, 1, "", "num_medusa_heads"], [87, 11, 1, "", "paged_state"], [87, 11, 1, "", "quant_mode"], [87, 11, 1, "", "redrafter_draft_len_per_beam"], [87, 11, 1, "", "redrafter_num_beams"], [87, 11, 1, "", "remove_input_padding"], [87, 11, 1, "", "rnn_conv_dim_size"], [87, 11, 1, "", "rnn_head_size"], [87, 11, 1, "", "rnn_hidden_size"], [87, 11, 1, "", "skip_cross_attn_blocks"], [87, 11, 1, "", "skip_cross_kv"], [87, 11, 1, "", "state_dtype"], [87, 11, 1, "", "state_size"], [87, 11, 1, "", "tokens_per_block"], [87, 11, 1, "", "trtllm_modules_to_hf_modules"], [87, 11, 1, "", "vocab_size"]], "tensorrt_llm.runtime.ModelRunner": [[87, 13, 1, "", "dtype"], [87, 12, 1, "", "from_dir"], [87, 12, 1, "", "from_engine"], [87, 13, 1, "", "gather_context_logits"], [87, 13, 1, "", "gather_generation_logits"], [87, 12, 1, "", "generate"], [87, 13, 1, "", "hidden_size"], [87, 13, 1, "", "mapping"], [87, 13, 1, "", "max_prompt_embedding_table_size"], [87, 13, 1, "", "max_sequence_length"], [87, 13, 1, "", "num_heads"], [87, 13, 1, "", "num_layers"], [87, 13, 1, "", "remove_input_padding"], [87, 12, 1, "", "serialize_engine"], [87, 13, 1, "", "use_lora_plugin"], [87, 13, 1, "", "vocab_size"], [87, 13, 1, "", "vocab_size_padded"]], "tensorrt_llm.runtime.ModelRunnerCpp": [[87, 13, 1, "", "dtype"], [87, 12, 1, "", "from_dir"], [87, 13, 1, "", "gather_context_logits"], [87, 13, 1, "", "gather_generation_logits"], [87, 12, 1, "", "generate"], [87, 13, 1, "", "hidden_size"], [87, 13, 1, "", "max_prompt_embedding_table_size"], [87, 13, 1, "", "max_sequence_length"], [87, 13, 1, "", "num_heads"], [87, 13, 1, "", "num_layers"], [87, 13, 1, "", "remove_input_padding"], [87, 13, 1, "", "vocab_size"], [87, 13, 1, "", "vocab_size_padded"]], "tensorrt_llm.runtime.MultimodalModelRunner": [[87, 13, 1, "", "audio_engine_dir"], [87, 13, 1, "", "cpp_e2e"], [87, 13, 1, "", "cpp_llm_only"], [87, 12, 1, "", "generate"], [87, 12, 1, "", "get_audio_features"], [87, 12, 1, "", "get_rope_index"], [87, 12, 1, "", "get_visual_features"], [87, 12, 1, "", "init_audio_encoder"], [87, 12, 1, "", "init_image_encoder"], [87, 12, 1, "", "init_llm"], [87, 12, 1, "", "init_processor"], [87, 12, 1, "", "init_tokenizer"], [87, 13, 1, "", "llm_engine_dir"], [87, 12, 1, "", "load_test_audio"], [87, 12, 1, "", "load_test_data"], [87, 12, 1, "", "prepare_position_ids_for_cogvlm"], [87, 12, 1, "", "preprocess"], [87, 12, 1, "", "ptuning_setup"], [87, 12, 1, "", "ptuning_setup_fuyu"], [87, 12, 1, "", "ptuning_setup_llava_next"], [87, 12, 1, "", "ptuning_setup_phi3"], [87, 12, 1, "", "ptuning_setup_pixtral"], [87, 13, 1, "", "python_e2e"], [87, 12, 1, "", "run"], [87, 12, 1, "", "setup_fake_prompts"], [87, 12, 1, "", "setup_fake_prompts_qwen2vl"], [87, 12, 1, "", "setup_fake_prompts_vila"], [87, 12, 1, "", "setup_inputs"], [87, 12, 1, "", "split_prompt_by_images"], [87, 12, 1, "", "tokenizer_image_token"], [87, 12, 1, "", "video_preprocess"], [87, 13, 1, "", "visual_engine_dir"]], "tensorrt_llm.runtime.QWenForCausalLMGenerationSession": [[87, 12, 1, "", "generate"]], "tensorrt_llm.runtime.SamplingConfig": [[87, 11, 1, "", "bad_words_list"], [87, 11, 1, "", "beam_search_diversity_rate"], [87, 11, 1, "", "early_stopping"], [87, 11, 1, "", "end_id"], [87, 11, 1, "", "frequency_penalty"], [87, 11, 1, "", "length_penalty"], [87, 11, 1, "", "max_attention_window_size"], [87, 11, 1, "", "max_new_tokens"], [87, 11, 1, "", "min_length"], [87, 11, 1, "", "min_p"], [87, 11, 1, "", "no_repeat_ngram_size"], [87, 11, 1, "", "num_beams"], [87, 11, 1, "", "num_return_sequences"], [87, 11, 1, "", "output_cum_log_probs"], [87, 11, 1, "", "output_log_probs"], [87, 11, 1, "", "output_sequence_lengths"], [87, 11, 1, "", "pad_id"], [87, 11, 1, "", "presence_penalty"], [87, 11, 1, "", "random_seed"], [87, 11, 1, "", "repetition_penalty"], [87, 11, 1, "", "return_dict"], [87, 11, 1, "", "sink_token_length"], [87, 11, 1, "", "stop_words_list"], [87, 11, 1, "", "temperature"], [87, 11, 1, "", "top_k"], [87, 11, 1, "", "top_p"], [87, 11, 1, "", "top_p_decay"], [87, 11, 1, "", "top_p_min"], [87, 11, 1, "", "top_p_reset_ids"], [87, 12, 1, "", "update"], [87, 11, 1, "", "use_beam_hyps"]], "tensorrt_llm.runtime.Session": [[87, 13, 1, "", "context"], [87, 13, 1, "", "context_mem_size"], [87, 13, 1, "", "engine"], [87, 12, 1, "", "from_engine"], [87, 12, 1, "", "from_serialized_engine"], [87, 12, 1, "", "infer_shapes"], [87, 12, 1, "", "run"], [87, 13, 1, "", "runtime"], [87, 12, 1, "", "set_shapes"]], "tensorrt_llm.runtime.TensorInfo": [[87, 11, 1, "", "dtype"], [87, 11, 1, "", "name"], [87, 12, 1, "", "numel"], [87, 11, 1, "", "shape"], [87, 12, 1, "", "squeeze"], [87, 12, 1, "", "view"]], "trtllm-serve-disaggregated": [[30, 17, 1, "cmdoption-trtllm-serve-disaggregated-c", "--config_file"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated-r", "--request_timeout"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated-t", "--server_start_timeout"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated-c", "-c"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated-r", "-r"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated-t", "-t"]], "trtllm-serve-disaggregated_mpi_worker": [[30, 17, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", "--config_file"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-log_level", "--log_level"], [30, 17, 1, "cmdoption-trtllm-serve-disaggregated_mpi_worker-c", "-c"]], "trtllm-serve-serve": [[30, 17, 1, "cmdoption-trtllm-serve-serve-backend", "--backend"], [30, 17, 1, "cmdoption-trtllm-serve-serve-cluster_size", "--cluster_size"], [30, 17, 1, "cmdoption-trtllm-serve-serve-ep_size", "--ep_size"], [30, 17, 1, "cmdoption-trtllm-serve-serve-extra_llm_api_options", "--extra_llm_api_options"], [30, 17, 1, "cmdoption-trtllm-serve-serve-gpus_per_node", "--gpus_per_node"], [30, 17, 1, "cmdoption-trtllm-serve-serve-host", "--host"], [30, 17, 1, "cmdoption-trtllm-serve-serve-kv_cache_free_gpu_memory_fraction", "--kv_cache_free_gpu_memory_fraction"], [30, 17, 1, "cmdoption-trtllm-serve-serve-log_level", "--log_level"], [30, 17, 1, "cmdoption-trtllm-serve-serve-max_batch_size", "--max_batch_size"], [30, 17, 1, "cmdoption-trtllm-serve-serve-max_beam_width", "--max_beam_width"], [30, 17, 1, "cmdoption-trtllm-serve-serve-max_num_tokens", "--max_num_tokens"], [30, 17, 1, "cmdoption-trtllm-serve-serve-max_seq_len", "--max_seq_len"], [30, 17, 1, "cmdoption-trtllm-serve-serve-num_postprocess_workers", "--num_postprocess_workers"], [30, 17, 1, "cmdoption-trtllm-serve-serve-port", "--port"], [30, 17, 1, "cmdoption-trtllm-serve-serve-pp_size", "--pp_size"], [30, 17, 1, "cmdoption-trtllm-serve-serve-reasoning_parser", "--reasoning_parser"], [30, 17, 1, "cmdoption-trtllm-serve-serve-tokenizer", "--tokenizer"], [30, 17, 1, "cmdoption-trtllm-serve-serve-tp_size", "--tp_size"], [30, 17, 1, "cmdoption-trtllm-serve-serve-trust_remote_code", "--trust_remote_code"], [30, 17, 1, "cmdoption-trtllm-serve-serve-arg-MODEL", "MODEL"]]}, "objnames": {"0": ["c", "macro", "C macro"], "1": ["cpp", "type", "C++ type"], "2": ["cpp", "class", "C++ class"], "3": ["cpp", "function", "C++ function"], "4": ["cpp", "functionParam", "C++ function parameter"], "5": ["cpp", "member", "C++ member"], "6": ["cpp", "enum", "C++ enum"], "7": ["cpp", "enumerator", "C++ enumerator"], "8": ["cpp", "templateParam", "C++ template parameter"], "9": ["py", "module", "Python module"], "10": ["py", "class", "Python class"], "11": ["py", "attribute", "Python attribute"], "12": ["py", "method", "Python method"], "13": ["py", "property", "Python property"], "14": ["py", "function", "Python function"], "15": ["py", "pydantic_field", "Python field"], "16": ["py", "pydantic_validator", "Python validator"], "17": ["std", "cmdoption", "program option"]}, "objtypes": {"0": "c:macro", "1": "cpp:type", "2": "cpp:class", "3": "cpp:function", "4": "cpp:functionParam", "5": "cpp:member", "6": "cpp:enum", "7": "cpp:enumerator", "8": "cpp:templateParam", "9": "py:module", "10": "py:class", "11": "py:attribute", "12": "py:method", "13": "py:property", "14": "py:function", "15": "py:pydantic_field", "16": "py:pydantic_validator", "17": "std:cmdoption"}, "terms": {"": [0, 1, 2, 3, 4, 6, 7, 8, 12, 14, 15, 16, 17, 18, 19, 20, 21, 23, 24, 26, 27, 28, 29, 31, 45, 46, 50, 51, 52, 58, 65, 69, 70, 71, 73, 75, 77, 78, 79, 80, 82, 83, 84, 87, 88, 89, 90, 92, 93, 95, 96, 97, 98], "0": [0, 1, 2, 3, 5, 6, 7, 9, 10, 12, 13, 15, 16, 17, 19, 20, 22, 23, 25, 26, 27, 28, 29, 30, 33, 34, 35, 36, 37, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 56, 57, 58, 59, 60, 62, 64, 65, 66, 67, 69, 70, 71, 72, 73, 74, 75, 79, 80, 81, 82, 83, 84, 87, 88, 89, 91, 92, 94, 95, 99], "00": [16, 26, 55, 56, 57, 73, 74, 75, 92], "000": [20, 73], "0000": [73, 75], "0007503032684326172": 30, "001": 51, "0012": 73, "0017": 74, "003": 74, "0047": 92, "005": 74, "0070": 92, "0071": 92, "0096": 92, "00978": 90, "01": [25, 26, 55, 56, 57, 73, 74, 89, 93], "014": 23, "0158": 75, "016": 74, "0162": 77, "0165": 79, "017": 74, "02": [74, 93], "021": 74, "022": 74, "0235": 92, "0260": 92, "0273": 92, "028": 74, "0294": 92, "03": [79, 92, 93], "032": 26, "0339": 74, "03762": 82, "03961": 4, "03x": 27, "04": [66, 67, 74, 91, 93, 94], "043": 74, "0449": 92, "0461": 20, "0463": 74, "05": [74, 82, 83, 84, 92, 93], "05100": 82, "0523": 92, "055": 74, "0554": 75, "0560": 92, "0563": 74, "06": [26, 73, 74, 82, 83], "0630": 92, "0669": 20, "068": 74, "0682": 92, "0689e": 73, "07": [25, 26, 74, 93], "0704": 75, "0713": 92, "0723": 92, "0732": 92, "0758": 20, "0772": 20, "0776": 92, "08": [26, 74, 79], "0804": 92, "082": 74, "0838": 74, "0881": 80, "089": 74, "09": [26, 92], "0903": 92, "0910": 92, "092": 74, "09353": 10, "0964": 74, "09685": 10, "097": 74, "09f": [0, 1], "0b": 2, "0e": 6, "0f": [0, 6, 70], "0rc1": 73, "0u": 1, "0x": 22, "0x0000000000000000": 93, "1": [0, 1, 2, 3, 5, 6, 7, 9, 10, 11, 12, 13, 15, 17, 19, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 33, 35, 36, 37, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 58, 59, 60, 62, 64, 66, 67, 69, 70, 72, 73, 75, 76, 77, 79, 81, 82, 83, 84, 86, 87, 88, 91, 92, 94, 98], "10": [0, 9, 10, 12, 20, 25, 26, 27, 30, 36, 38, 43, 44, 51, 59, 64, 67, 70, 73, 74, 75, 77, 80, 82, 89, 91, 92], "100": [0, 9, 20, 30, 38, 56, 72, 73, 75, 88], "1000": [0, 72, 73, 74, 75], "10000": [82, 83, 84], "1003": 93, "100gb": 28, "101": 9, "101230": 51, "101978": 74, "102": [9, 22], "1024": [1, 6, 15, 20, 23, 25, 29, 36, 43, 44, 51, 54, 70, 73, 74, 75, 79, 82, 83, 92], "103": 9, "104": 93, "10438": 90, "1045": 92, "1047": 73, "1050": 92, "1051": 75, "1059": 73, "106563": 74, "1072": 92, "107501": 74, "10764": 53, "10774": 0, "1079": 19, "108": 74, "1082": 92, "10858": 36, "10b": [69, 82, 93], "10m": 22, "11": [0, 10, 12, 20, 23, 25, 64, 73, 74, 77, 82, 92], "11023": 73, "110804": 74, "110b": 93, "111": [22, 26], "111302": 74, "111618": 74, "111668": 74, "1118": 93, "1123": 93, "1134": 89, "1135": 92, "1141": 92, "1148": 93, "11489": 20, "11490": 73, "1151": 20, "115716": 74, "1160": [30, 37], "117": 74, "1178": 73, "1181": 93, "1183": 93, "119": 73, "11943": 73, "11947": 36, "1196": 20, "11b": [91, 93], "12": [0, 10, 15, 22, 26, 36, 64, 66, 67, 73, 74, 77, 79, 82, 92], "1207": 53, "1212": 92, "121847": 73, "1219": 20, "122": 73, "1225": 82, "12288": 73, "123": [30, 38, 39], "1234": [70, 84], "1239": 93, "1242": 93, "1248": 93, "125": 73, "1252": [19, 73], "1256": 93, "125m": [12, 15], "126": 73, "1267": 93, "127": 82, "1272": 92, "128": [0, 1, 5, 9, 10, 13, 16, 20, 21, 22, 23, 24, 25, 26, 30, 36, 38, 39, 49, 56, 70, 73, 74, 93], "1284": 93, "1287": 77, "1290": 92, "1291504": 75, "1293": 19, "12945": 20, "129498": 20, "13": [5, 10, 24, 28, 64, 73, 74, 75, 82, 92], "1300": 45, "13044": 53, "131072": [73, 75], "13195": 73, "132": [73, 74], "1323": 93, "1328": 93, "1329": 93, "133": 93, "13368": 73, "1337": 93, "1341": 20, "1343": 93, "1344": 93, "13525": 73, "13598": 73, "1363": 53, "137": 73, "1378": 92, "139": 74, "1392": 93, "13b": 22, "14": [10, 15, 25, 64, 73, 74, 77, 79, 80, 92], "140g": 19, "141": 23, "1418": 73, "141gb": [21, 74], "142": 28, "1424": 93, "1436": [20, 93], "1437": 92, "144": 77, "1446": 93, "1447": 93, "14480": 73, "1449": 93, "145": [79, 80], "1459": 92, "146": [79, 80], "1467": 93, "147": [75, 77, 79, 80], "1480": 93, "1486": 93, "149": [92, 93], "15": [10, 26, 64, 73, 74, 80, 82, 92], "150": 72, "1500": 74, "15043": 36, "1514": 93, "1529": 93, "1534": 93, "1535": 93, "1536": 20, "1537": 93, "1539": 93, "154": 26, "1552": 93, "1556": 92, "15585": 73, "1562": 93, "1564": [75, 79, 80], "158": 20, "1583": 93, "1584": 20, "1585": 75, "15889": 53, "1589": 93, "1590": 93, "1597": 77, "15u": 28, "16": [0, 5, 10, 11, 12, 16, 20, 22, 25, 26, 30, 33, 35, 55, 56, 57, 64, 65, 73, 74, 75, 76, 82, 83, 84, 89, 90, 92], "160": 93, "1607": 73, "161": [30, 37, 73], "1625": 77, "1626": 93, "163": 21, "1637": 93, "16384": [77, 79], "164": 26, "1642": 93, "1650": 93, "1660": 93, "1669": 93, "167": [73, 74], "1672": 92, "1674": 93, "1675": 93, "1676": 93, "168": 26, "16e": 91, "16x": [27, 89], "17": [0, 2, 10, 20, 64, 73, 74, 79, 92, 94], "1706": 82, "1721": 92, "1723": 93, "17233": 20, "173": 26, "1732": 93, "17323": 90, "1738": 93, "174": 74, "1741966075": 88, "1742": 93, "17453": 29, "17453v3": 1, "175": 74, "175b": 23, "176": 73, "176064": 20, "1762": 93, "1799": 93, "17b": 91, "18": [2, 10, 28, 64, 71, 73, 74, 92], "180": [26, 89], "180000000": 0, "180b": [25, 73], "1811": 53, "1815": 93, "181540": 20, "182": 74, "1822": 36, "183": 74, "1834": 93, "184": 74, "185": [22, 73], "1851": 93, "18527": 36, "18533": 53, "18563": 73, "1861": 80, "1866": 80, "1885": 75, "1886": 93, "1889": 53, "1897": 93, "19": [2, 20, 64, 74, 80, 92], "1900": 53, "1909": 93, "191": 74, "192": 21, "1921": 20, "1926": 93, "1937": 93, "1939": 93, "1944": 79, "1953": 93, "1959": 73, "198": 26, "1985": 93, "1987": 93, "1993": 92, "1999": 93, "1_405b": 16, "1_70b": 16, "1b": [30, 33, 35, 38, 40, 42, 45, 46, 47, 48, 49, 50, 51, 52, 53, 58, 59, 60, 62, 66, 67, 69, 88, 94], "1d": [5, 82, 87], "1e": [15, 82, 83, 84], "1e20f": 1, "1g": 92, "1gb": 2, "1k": [20, 26, 27, 28], "1m": 80, "1st": [22, 82, 89], "1u": [0, 1], "1x": 26, "1xh200": 21, "1ytic": 93, "2": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15, 16, 19, 21, 22, 23, 25, 26, 27, 28, 30, 42, 44, 48, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 64, 66, 67, 69, 70, 73, 74, 76, 77, 79, 80, 82, 84, 87, 90, 91, 92, 98], "20": [1, 6, 12, 13, 28, 30, 60, 62, 73, 74, 75, 79, 82, 87, 92], "200": [23, 70, 87], "2000": [28, 74], "20000": 74, "200mb": 28, "2017": 79, "2018": 93, "2023": [21, 92], "2024": 26, "2025": [20, 26, 73], "2028": 93, "203": 74, "2033": 80, "2039": 93, "204": [26, 74], "2040": 93, "2044": [79, 80], "2045": 79, "2048": [15, 20, 21, 23, 24, 28, 29, 49, 70, 73, 74, 75, 77, 78, 79, 80, 84, 87, 92, 93], "2056": 93, "206": 74, "20627": 36, "20685": 73, "2079": 92, "208": 74, "2081": [77, 79, 93], "2087": 93, "2089": 74, "209": 74, "20b": 93, "21": [12, 25, 26, 74, 79, 92, 93], "2101": 4, "2102": 74, "2106": 10, "2107": [53, 92], "210g": 19, "211": 26, "2113": 93, "2135": 93, "21367": 53, "2152": 93, "2158": 74, "2168": 20, "2169": 93, "21747": 73, "2176": 74, "21764": 73, "2182": 93, "2191": 93, "22": [28, 32, 74, 82, 92], "22000": 74, "22056": 73, "221": 73, "2210": 90, "2211": [82, 90], "2219": 93, "22213": 73, "2225": 92, "2232": 93, "224": 83, "2243": 93, "2263": 93, "227": 24, "2288": 93, "2294": 93, "22x": 27, "23": [73, 74, 92, 93], "2305": 92, "2306": 90, "2309": [1, 29], "232": 24, "2337": 53, "2352": 93, "2357": 93, "236": 26, "2366": 93, "2370": 93, "2373": 93, "2379": 93, "2388": 93, "239": 26, "2397": 73, "24": [0, 66, 67, 74, 92, 93, 94], "240": 74, "2401": 0, "2402": 10, "24189": 74, "2419": 93, "242": 74, "2425": 93, "2439": 93, "245": 26, "2458": 93, "2461": 79, "2466": 79, "2473": 93, "2474": [77, 79], "2484": 93, "2485": 93, "2487": 74, "249": 26, "25": [24, 26, 73, 74, 91, 93], "250": [20, 26], "2500": 74, "25032": 73, "252u": 28, "253": [26, 74], "2552": 93, "256": [1, 20, 21, 24, 28, 59, 70, 73, 74, 82, 92, 93], "25603": 73, "2573": 93, "2581": [77, 79], "2590780": 73, "259840": 89, "26": [73, 74, 77, 88], "260": 74, "2602": 36, "2628": [79, 80], "263": [21, 36, 53], "2640": 80, "2649": 92, "2671": 20, "2677": 93, "26778": 73, "2679": 77, "2685": 93, "2688": 53, "2691": 93, "27": [74, 93], "270": 74, "2712": 93, "274": [20, 93], "2742": 75, "275": 93, "27556": 53, "276": 74, "278": [36, 53, 74], "2782": 93, "2787": 93, "2796": 93, "28": [26, 73, 74, 92], "2820": 92, "2826700": 20, "28390": 73, "287113": 73, "288": 93, "29": [74, 89], "292": 74, "2939": 92, "294": 74, "297": 36, "29889": 53, "29892": 36, "299": [26, 73], "29962": 36, "2998": 92, "2b": [19, 64, 73], "2cta": 28, "2d": [12, 82, 83, 90], "2k": [20, 26, 27, 28], "2m": 80, "2nd": 82, "2u": 1, "2x": [22, 23], "3": [0, 1, 3, 5, 7, 9, 10, 17, 21, 22, 23, 25, 26, 27, 28, 43, 44, 46, 48, 52, 54, 58, 59, 64, 66, 67, 69, 70, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 87, 88, 92, 93, 94, 95], "30": [0, 12, 20, 26, 70, 74, 75, 77, 80, 82, 89], "300": [24, 73], "3000": [73, 74], "30000": 74, "30065": 73, "3019": 73, "3021": 20, "3022": 73, "303": 23, "3031": 79, "304": [36, 53], "3040": [75, 79, 80], "306": 36, "3072": 20, "30990": 73, "30b": 25, "30x": 25, "31": [11, 74, 75, 79, 80], "311": 74, "3132": 73, "315": [26, 74], "318": 74, "32": [1, 5, 9, 11, 20, 22, 23, 29, 36, 53, 70, 73, 74, 75, 82, 83, 84, 87, 88, 89, 90, 92, 93, 94], "3201": 75, "321": 73, "322": [36, 53], "3276": [75, 79, 80], "32768": 82, "3291": 92, "32b": 93, "32k": 93, "32x": 25, "33": [74, 92], "332": 74, "3328": 92, "3338": 75, "338": [26, 36, 53], "3389": 77, "33x": 27, "34": [20, 74], "340": [26, 74], "341": 23, "3442": 92, "3445": 92, "3452": 92, "3476": 20, "349": 23, "34b": 93, "35": [0, 70, 74], "351": 74, "3555": 92, "35611": 20, "357": 74, "36": [26, 74, 76, 77], "3671": 73, "368": 26, "37": 73, "370": 74, "371": 74, "374": 74, "375": 74, "3763": 26, "379": 74, "38": [73, 74], "384": [20, 74], "3863": 74, "387": 74, "387b12598a9e": 73, "3885": 20, "3887": 92, "39": [26, 74], "3914": 74, "3936": 73, "3977": 92, "399": 74, "3_1": 91, "3_3": 91, "3b": [34, 39, 61], "3d": [5, 82, 87], "3rd": 82, "3u": 1, "3x": [25, 26, 28], "4": [0, 1, 2, 7, 9, 10, 11, 12, 16, 19, 23, 25, 26, 27, 28, 30, 36, 43, 44, 49, 53, 54, 55, 56, 57, 64, 70, 73, 74, 75, 77, 78, 79, 80, 81, 82, 84, 87, 88, 89, 90, 91, 92, 93], "40": [6, 74, 77, 82, 93], "400": 28, "4000": 28, "403": 93, "405": 53, "405b": [73, 76], "4060": 89, "4066": 36, "408": 74, "4089": 80, "4096": [21, 28, 36, 73, 74, 77, 82, 83, 87], "40b": 25, "40gb": 29, "40x": 25, "41": 74, "41020": 73, "411": 73, "4117e": 73, "4133": 80, "41375": 73, "414": 20, "41607": 73, "4168": 20, "4192": 92, "42": [52, 73, 74], "4203099703668305365": 51, "4224": 74, "4248": 77, "4265": 73, "427": [53, 73, 74], "4280": 26, "43": [74, 88, 89], "433": 74, "437": 74, "438": 74, "44": [74, 89], "4408": 36, "442": 74, "4439": 73, "4451": 20, "4456": 74, "447": 74, "448": 74, "449": 93, "4493": [20, 79, 80], "4497": 74, "44x": 25, "45": [9, 74, 91, 93], "450": 74, "45000000000": 9, "453": 74, "4566": 74, "459": 74, "46": 25, "4600": 28, "462": 74, "463": 74, "4653": 36, "4656": 74, "466": 74, "4667": 74, "47": [25, 77], "4701": 73, "471": 74, "472": 36, "475": 74, "477": 74, "478": 93, "47x": 25, "48": [74, 77, 89, 93], "481": [22, 74], "482": 93, "488": 74, "49": [74, 77], "49152": 20, "495": 74, "496": 11, "4963": 73, "49b": 91, "4b": 93, "4bit": 21, "4gb": 28, "4u": 1, "4x": [21, 22, 23], "5": [0, 1, 9, 10, 12, 13, 15, 21, 22, 23, 25, 26, 27, 28, 34, 39, 44, 45, 51, 54, 61, 69, 70, 73, 74, 79, 82, 84, 87, 91, 92, 93], "50": [0, 25, 45, 70, 73, 74, 93], "500": [26, 28, 74], "5000": 74, "500000": 84, "5001": 53, "5007": 36, "500m": 25, "50272": 15, "505143404006958": 30, "5064": 74, "5073": 92, "51": 74, "512": [1, 10, 13, 23, 24, 70, 73, 74, 77, 79, 84], "5120": 20, "512mb": 2, "514": 74, "518": [36, 74], "51b": [91, 93], "51x": 25, "52269": 74, "524": 74, "525": 74, "526": [53, 74, 93], "52667": 74, "529": 74, "5299": 77, "53": [73, 79, 80], "5305": 77, "531": 74, "54": [25, 74], "540": 73, "543": 74, "544": 74, "5496": 77, "5497": 74, "55": [25, 73, 74], "5500": 74, "5510": 73, "5514": 73, "5530": 74, "554": 74, "557": 74, "559": 74, "56": [25, 74], "560": 21, "562": [10, 13], "56401920000": 30, "565": 74, "567": 74, "568": [73, 74], "57": [73, 74], "571": 74, "572": 74, "5739": 20, "5742": [77, 79], "579": 74, "58": [26, 74, 79], "580": 74, "5821": 74, "5830": 92, "5874": 92, "5877": 77, "5879": 92, "588": 74, "58x": 26, "59": 73, "590": [36, 74], "5918": 92, "5942": 20, "5957": 92, "5976": 77, "598": 74, "5980": 77, "5b": 93, "5th": [28, 82], "5u": 1, "5x": [22, 25, 26], "6": [0, 1, 6, 9, 10, 12, 23, 25, 26, 27, 28, 30, 44, 54, 70, 74, 82, 87, 91, 92, 93], "60": [0, 74], "600": 31, "6000": 73, "602": 74, "6049": 77, "6059": 73, "6064": 92, "608": 74, "61": 74, "610": 74, "6100": 20, "6157": 92, "618": 74, "62": [26, 74, 79], "6255": 92, "626": 36, "6299": 92, "63": [43, 44, 54, 65, 73, 74, 79, 84, 89], "630": 74, "63266": 75, "63307": 75, "63308": 75, "63331": 75, "63374": 75, "634": 74, "63456": 75, "6345624": 75, "6372": 77, "639": 93, "64": [0, 1, 5, 6, 15, 20, 22, 23, 29, 34, 39, 58, 61, 74, 79, 82, 83, 84, 89, 93], "640": [21, 74], "640gb": 28, "6452": 80, "6475": 79, "649": 93, "64x": 26, "65": [67, 74], "65024": 92, "6523": 80, "653": 74, "654": 23, "6550": 77, "6554": 79, "656": 74, "657": 74, "659": 74, "6591": 73, "66": [26, 74], "661": 74, "6628": [79, 80], "6678": 89, "6684": 80, "6695": 89, "67": [25, 26, 74], "6701": 20, "671": 20, "67108864": 65, "671b": 27, "673": 93, "675": 73, "6753e": 73, "6769": 79, "679": 22, "68": [25, 26, 74, 80], "682": 74, "6825": 73, "683": 74, "684": 26, "685": 74, "6852": [77, 79], "686": 74, "6862": 73, "6890": 92, "69": [25, 26, 74, 80, 88], "6925": 73, "6938": 36, "695": 93, "696": 74, "697": 28, "6975": 77, "6976": [75, 79, 80], "698": 74, "6a": 21, "6b": [22, 73, 82, 93], "6x": 23, "7": [0, 1, 9, 10, 21, 22, 25, 26, 27, 28, 44, 54, 64, 65, 66, 67, 73, 74, 75, 82, 87, 92], "70": [0, 25, 80, 89], "700": 31, "7000": 73, "701": 93, "7031": 77, "704": 74, "705": [28, 93], "706": 74, "7063": 73, "707": 74, "7072": 74, "709": 73, "7090": 92, "70b": [5, 19, 23, 25, 54, 75, 77, 78, 79, 80, 81, 91, 93], "70g": 19, "71": [26, 73, 74], "711": 74, "712": 74, "7134": 92, "7136": 75, "714": 74, "7144": 92, "7168": [26, 28], "717": 74, "7187": 74, "7188": 20, "72": [74, 76], "722": 74, "727": 74, "72b": [91, 93], "73": [26, 74], "732": 74, "734": 74, "736": 74, "737": 74, "7382": 74, "739": 93, "74": [26, 74], "741": [74, 93], "742": 74, "745": 74, "7456": 20, "74561": 20, "747": 74, "7480": 75, "75": [25, 73, 93], "750": [23, 74], "7502": 75, "7520": 20, "755": 31, "7584": 20, "75903": 74, "76": 74, "7607": 79, "7621": 74, "7638": [75, 79, 80], "767": 74, "768": [15, 83], "77": 74, "772": 74, "7743": 75, "7770": 75, "78": [26, 74, 77], "780": 73, "7842": 77, "78509": 74, "7876": 79, "79": [73, 89], "7900": 92, "7933": 79, "794": [74, 93], "7949": 92, "7977": 77, "7a": 21, "7b": [10, 12, 13, 25, 30, 43, 44, 54, 73, 74, 88, 91, 93], "7x": [22, 26], "8": [0, 1, 5, 9, 10, 11, 15, 16, 19, 20, 21, 23, 24, 25, 26, 27, 28, 29, 30, 36, 37, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 55, 56, 57, 59, 64, 66, 67, 70, 73, 74, 75, 76, 77, 81, 82, 83, 84, 88, 89, 90, 92], "80": [0, 6, 23, 26, 28, 65, 74, 93], "800": [21, 74, 93], "8000": [30, 33, 34, 35, 37, 38, 39, 60, 61, 62, 88], "8002": 73, "8005": 74, "803": 21, "8048": 73, "80gb": [22, 25, 29, 74, 75, 77, 78], "81": [26, 74, 77], "810": 74, "8149": 92, "8179": 92, "819": 23, "8192": [29, 70, 73, 74, 75, 79, 82, 83, 92, 93], "82": [26, 74, 77], "820": 73, "8212": 1, "8218": 92, "822": 74, "8225": 77, "825": 93, "8259": 73, "83": 74, "8307": 80, "8351": 73, "838": 74, "84": [26, 74], "840": 74, "841": 74, "8441": 73, "85": [20, 25, 73, 74, 93], "850": 74, "851": 74, "854": 74, "86": [65, 74], "863": 73, "866": 74, "867": 74, "8672": 92, "87": [25, 74], "8779": 92, "88": [74, 77, 80], "8804": 75, "8828": 92, "8841": 77, "89": [25, 26, 65, 74, 91], "893": 74, "8932": 73, "8958": 80, "896": [53, 74], "8a": 24, "8b": [46, 54, 69, 73, 88, 91, 94], "8bit": 22, "8tb": 23, "8x": 28, "8x7b": [4, 73, 91, 93], "8xb200": 26, "8xgpu": 28, "8xh100": 24, "8xh200": 21, "9": [0, 1, 10, 12, 19, 22, 26, 27, 44, 54, 59, 64, 74, 77, 82, 92], "90": [0, 20, 65, 70, 73, 74, 75, 77, 81, 89], "9007": 20, "9028": 92, "907": 22, "9087": 80, "91": 74, "910": 74, "9101": 74, "911": 74, "9115": 80, "912656": 20, "913": 74, "9184": 77, "92": [26, 74], "920": 74, "9203": 77, "9214": 74, "924": 15, "925": 74, "9274": 75, "93": 74, "935": 93, "9353e": 75, "9379": 20, "94": 74, "94022": 74, "941": [21, 24], "943": 53, "944": 74, "946": 21, "947": 74, "9494": 79, "95": [30, 37, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 59, 66, 67, 74, 75, 81, 88], "9521": 92, "953": 74, "9537": 77, "954": 28, "956": 74, "957": 74, "96": [21, 26, 28, 74, 77, 93], "960": 21, "9606": 28, "961": 74, "9613": 28, "9623": 79, "9629": 28, "963": 74, "9639": 74, "96583": 74, "967": 93, "9692": 92, "97": [28, 73, 74, 77], "970": 74, "98": 74, "983": 93, "987": 93, "99": [9, 26, 31, 74], "990": 74, "991": 74, "992": 93, "9928": 80, "9938": 20, "9982": [79, 80], "9x": [23, 24], "A": [0, 1, 2, 3, 5, 6, 8, 10, 12, 15, 16, 19, 20, 25, 26, 52, 55, 56, 57, 58, 70, 72, 73, 74, 82, 87, 93, 95, 97], "AND": 82, "And": [12, 19, 27, 28, 82, 83, 89], "As": [4, 5, 7, 10, 12, 16, 18, 27, 36, 77, 80, 81, 82, 89, 90, 92, 97, 98], "At": [14, 28, 58, 77, 83, 89], "But": [5, 8, 71], "By": [0, 1, 2, 6, 12, 26, 28, 36, 65, 70, 73, 77, 80, 82, 92, 97], "For": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 12, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 36, 40, 52, 55, 56, 57, 63, 65, 69, 73, 74, 75, 76, 77, 79, 80, 81, 82, 87, 88, 89, 92, 93, 95, 96, 97, 98, 99], "If": [0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 15, 16, 17, 19, 25, 27, 29, 30, 31, 32, 65, 66, 67, 69, 70, 71, 73, 75, 76, 77, 79, 80, 81, 82, 84, 87, 89, 91, 92, 93, 95, 97, 98, 99], "In": [0, 1, 2, 7, 8, 11, 12, 16, 17, 19, 20, 22, 25, 26, 27, 28, 32, 36, 54, 58, 64, 65, 73, 74, 75, 76, 77, 79, 80, 82, 88, 89, 90, 91, 92, 93, 97, 98, 99], "It": [0, 1, 3, 5, 6, 7, 10, 12, 14, 16, 17, 18, 20, 21, 24, 25, 26, 27, 28, 29, 36, 51, 58, 65, 70, 71, 73, 74, 77, 78, 79, 80, 81, 82, 88, 90, 92, 95, 96, 97, 99], "Its": [5, 82, 97], "NOT": 82, "No": [0, 2, 9, 58, 73, 75], "Not": [1, 25], "ON": [73, 77, 79, 80], "OR": 82, "Of": [26, 93], "On": [5, 9, 65, 67, 72, 76, 80, 82, 93], "One": [2, 15, 16, 79, 82, 92, 96], "Or": [82, 87, 94], "That": [3, 5, 6, 9, 16, 71, 77, 82], "The": [0, 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 25, 26, 27, 28, 29, 30, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 52, 53, 54, 55, 56, 57, 58, 59, 64, 65, 66, 67, 69, 70, 72, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 84, 85, 87, 88, 89, 91, 92, 93, 94, 95, 96, 97, 98, 99], "Their": 28, "Then": [10, 19, 27, 30, 31, 73, 75, 82, 95, 98], "There": [2, 5, 6, 7, 8, 9, 10, 15, 19, 23, 26, 27, 28, 36, 65, 67, 69, 82, 85, 89, 90, 92, 93, 96, 97, 98, 99], "These": [2, 12, 19, 21, 23, 24, 26, 28, 36, 73, 75, 76, 83, 85, 88, 93], "To": [2, 3, 5, 9, 10, 12, 13, 16, 17, 18, 19, 20, 23, 26, 27, 65, 69, 70, 71, 72, 73, 74, 77, 79, 80, 81, 82, 88, 89, 90, 93, 94, 95, 97, 98, 99], "Will": 0, "With": [5, 6, 12, 16, 31, 36, 49, 64, 73], "_": [0, 3, 17, 85], "__all__": 95, "__call__": 52, "__init__": [7, 14, 16, 17, 52, 70, 73, 92, 93, 95, 97, 99], "__main__": [40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 66, 67, 69, 75, 77, 80, 81, 88, 93, 94, 95], "__name__": [40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 66, 67, 75, 77, 80, 81, 88, 93, 94, 95], "__post_init__": 93, "__repr__": 93, "_capac": 1, "_context_logits_auto_en": 70, "_cpp_gen": 3, "_create_tensor": 16, "_explicitly_disable_gemm_plugin": 85, "_generation_logits_auto_en": 70, "_handl": 1, "_mark_output": 92, "_mpi_sess": 70, "_note": 5, "_num_postprocess_work": 70, "_path": 20, "_postproc_param": 70, "_postprocess_result": 70, "_postprocess_tokenizer_dir": 70, "_reasoning_pars": 70, "_return_log_prob": 70, "_run": 92, "_runtim": 87, "_str_to_trt_dtype_dict": 82, "_torch": [73, 93, 94, 95, 96, 97], "_torchllmargs__context": 70, "_trtllmargs__context": 70, "_unsign": 1, "_util": 82, "a10": 29, "a100": [6, 19, 29], "a10g": 29, "a2": 93, "a30": 29, "a40": 29, "a8": 90, "a_": 82, "a_1": 82, "a_2": 82, "a_n": 82, "a_sf": 82, "aarch64": 91, "ab": [10, 29, 82, 90], "abbrevi": 30, "abc": 27, "abcd": 27, "abi": [65, 93], "abil": [71, 73], "abl": [5, 22, 26, 67, 73, 79, 82, 93], "ablat": [27, 28], "abnorm": 93, "abort": [70, 93], "about": [0, 1, 3, 19, 20, 21, 22, 24, 25, 28, 51, 58, 59, 64, 73, 75, 77, 78, 80, 82, 88, 89, 92, 93], "abov": [2, 10, 11, 16, 19, 20, 25, 28, 36, 65, 73, 74, 75, 77, 80, 89], "absenc": 6, "absorb": 26, "abstract": [80, 83], "ac": 93, "acc": 82, "acceler": [5, 11, 12, 22, 23, 24, 25, 29, 71], "accept": [0, 1, 12, 20, 36, 46, 47, 48, 49, 50, 65, 70, 75, 77, 82, 87, 88, 91, 93, 97], "accept_length": 87, "acceptancelength": 0, "acceptancer": 0, "acceptancethreshold": 0, "acceptedlen": 1, "acceptedlengthscumsum": 1, "acceptedpath": 1, "acceptedpathid": 1, "acceptedtoken": 1, "acceptedtokenslen": 1, "access": [3, 32, 45, 70, 73, 75, 82, 88, 93], "accessor": 1, "accommod": [4, 96, 98], "accomplish": 76, "accord": [5, 17, 59, 82, 83, 97], "accordingli": 17, "account": [16, 20, 31, 55, 56, 57, 65], "accumul": [0, 5, 6, 29, 70, 82, 87, 88], "accur": [21, 27, 45, 73, 75, 93], "accuraci": [21, 26, 28, 29, 77, 81, 82, 90, 93], "achiev": [2, 12, 20, 21, 25, 26, 28, 65, 74, 75, 77, 79, 81, 95], "across": [2, 4, 5, 6, 7, 16, 17, 23, 26, 30, 74, 76, 77, 79, 80, 82, 87], "act": 26, "act_fn": 83, "act_typ": [16, 82], "action": 54, "activ": [0, 1, 5, 7, 16, 20, 21, 22, 25, 26, 28, 29, 76, 82, 90, 91, 93, 99], "activation_scaling_factor": 15, "activationtyp": [16, 82], "active_request": 99, "actual": [7, 8, 12, 20, 25, 26, 27, 29, 77, 79, 80, 81, 93, 98], "ad": [1, 5, 6, 7, 9, 12, 13, 19, 27, 28, 32, 52, 64, 72, 76, 79, 80, 82, 84, 87, 93, 94, 96], "ada": [5, 25, 59, 65, 71, 77, 91, 93], "adalayernorm": 83, "adalayernormcontinu": 83, "adalayernormzero": 83, "adalayernormzerosingl": 83, "adapt": [0, 10, 27, 40, 41, 70, 82, 83, 93, 95], "adapter_s": 10, "adapters": 1, "add": [1, 3, 5, 7, 10, 14, 15, 16, 19, 27, 31, 32, 52, 54, 65, 69, 70, 73, 75, 77, 80, 82, 87, 92, 93, 95, 98], "add_activ": 16, "add_argu": 54, "add_bias_linear": 84, "add_generation_prompt": 26, "add_input": 82, "add_output": 82, "add_padding_request": 98, "add_qkv_bia": 84, "add_rmsnorm": 26, "add_sequ": 87, "add_special_token": [26, 70, 87, 93], "addcumlogprob": 93, "added_kv_proj_dim": 83, "added_proj_bia": 83, "addit": [0, 5, 6, 10, 12, 16, 19, 23, 27, 28, 30, 36, 45, 65, 70, 73, 74, 76, 77, 79, 82, 83, 90, 91, 92, 93, 97, 98], "addition": [2, 73, 75, 77, 80, 95, 97], "additional_model_output": 70, "additional_opt": 57, "additionalmodeloutput": [0, 3, 70], "additionaloutput": [0, 3], "addr": 0, "address": [1, 17, 20, 25, 26, 28, 69, 80, 89, 93], "addresswiths": 1, "adequ": 83, "adher": 45, "adjust": [55, 70, 73, 75, 89, 99], "admin": 67, "adopt": [6, 19], "advanc": [12, 16, 24, 26, 27, 28, 29, 42, 46, 47, 49, 50, 65, 70, 82, 93, 97], "advantag": [6, 71], "advers": [21, 29], "advertis": 73, "advis": 2, "affect": [11, 19, 20, 29, 75, 77, 79, 80, 89], "affin": 83, "after": [0, 1, 3, 5, 7, 8, 9, 10, 12, 16, 17, 26, 27, 28, 29, 30, 31, 51, 54, 65, 69, 70, 73, 77, 79, 80, 81, 82, 83, 85, 88, 89, 93, 97, 99], "again": [16, 75, 77, 80, 92], "against": [65, 73], "agent": 23, "agentdesc": 0, "agentnam": 0, "agentst": 0, "aggreg": 28, "aggress": [15, 27, 77, 81], "agre": [69, 88], "ahead": [0, 5, 12], "ai": [20, 22, 26, 30, 37, 40, 42, 43, 44, 46, 47, 48, 49, 50, 54, 59, 66, 67, 71, 72, 75, 81, 82, 88, 91, 93, 94], "aidc": 93, "aim": [4, 15, 20, 26, 71, 73, 75, 77, 93], "ainsli": 21, "air": 93, "aka": 82, "akhoroshev": 93, "al": 21, "albeit": 12, "alessionetti": 93, "algorithm": [0, 5, 6, 12, 15, 16, 19, 25, 26, 27, 28, 70, 73, 77, 82, 93], "alia": [70, 83, 84], "alibi": 82, "alibi_bias_max": [82, 83], "alibi_scal": 82, "alibi_slop": 82, "alibi_with_scal": 82, "align": [73, 93, 99], "align_corn": 82, "all": [0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 16, 17, 19, 20, 23, 26, 27, 28, 52, 55, 56, 57, 58, 65, 70, 71, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 85, 87, 88, 89, 90, 91, 92, 93, 97, 98, 99], "all_reduce_param": [82, 83], "allbitset": [0, 1], "allgath": [16, 28, 29, 80, 82, 93], "allgeneratedtoken": 0, "alllayersdrafttokenid": 1, "alllayersdrafttokenidspredecessor": 1, "alllayersscor": 1, "alloc": [0, 1, 2, 5, 8, 9, 30, 36, 70, 81, 82, 87, 89, 92, 93, 96, 97, 98, 99], "allocateipcmemori": 1, "allocatespeculativedecodingbuff": 1, "allocnewblock": 0, "allocnewblocksperrequest": 0, "alloctotalblock": 0, "alloctotalblocksperrequest": 0, "allot": 0, "allottedtimem": [0, 93], "allow": [0, 1, 2, 3, 5, 6, 9, 12, 15, 21, 24, 28, 29, 69, 70, 71, 72, 73, 74, 75, 76, 77, 79, 80, 82, 85, 92, 93, 96, 99], "allowed_token_id": 52, "allreduc": [16, 26, 28, 29, 80, 82, 93], "allreducebuff": 1, "allreducefusionkernel": 26, "allreducefusionop": 82, "allreduceparam": [82, 83], "allreducestrategi": [11, 82], "almost": [16, 28, 77, 79, 89], "alon": 4, "along": [5, 12, 18, 65, 82, 93], "alpaca": 10, "alpha": [70, 82, 83, 93], "alphabet": 82, "alreadi": [0, 5, 7, 9, 18, 20, 26, 27, 28, 70, 77, 79, 81, 82, 93, 95, 98], "also": [0, 2, 3, 5, 7, 12, 15, 16, 17, 18, 19, 20, 23, 24, 25, 26, 27, 28, 29, 30, 36, 49, 51, 52, 65, 69, 70, 73, 74, 75, 76, 77, 78, 79, 82, 83, 88, 89, 90, 93, 95, 96, 97, 98], "altair": 93, "alter": [3, 7], "altern": [3, 26, 52, 65, 95, 96], "although": [7, 16, 73, 77, 80], "alwai": [0, 1, 3, 5, 6, 9, 15, 16, 19, 28, 53, 70, 79, 80, 82, 92], "always_share_across_beam": 87, "am": [42, 46, 47, 49, 50, 52, 59, 75, 81, 87], "ambigu": 1, "amd": 93, "amen": [0, 3, 70], "among": [32, 82], "amongst": 82, "amount": [0, 9, 16, 28, 29, 70, 73, 79, 81, 87, 89, 92], "amper": [22, 65, 71, 91, 93], "an": [0, 1, 2, 3, 5, 6, 7, 9, 10, 12, 13, 14, 15, 16, 17, 18, 19, 21, 23, 25, 26, 27, 28, 29, 30, 36, 42, 45, 46, 47, 48, 49, 50, 52, 59, 65, 67, 69, 70, 71, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 87, 88, 89, 90, 92, 93, 95, 96, 97, 98, 99], "analysi": [7, 26, 27, 28, 64, 89], "analysispatternmanag": 7, "analyt": 22, "analyz": [7, 75], "ani": [0, 1, 2, 3, 7, 8, 12, 17, 19, 20, 28, 30, 52, 65, 69, 70, 71, 73, 74, 79, 80, 81, 82, 84, 87, 92, 95, 96, 97], "announc": [20, 21, 22, 24], "anoth": [0, 1, 5, 7, 10, 19, 22, 26, 27, 28, 30, 79, 82, 92, 97, 99], "answer": [27, 45], "antialia": 82, "antonin": [42, 46, 47, 49, 50], "anybitset": [0, 1], "anyth": [58, 74], "aotman": 93, "apart": 36, "api": [2, 6, 9, 12, 14, 15, 16, 18, 20, 27, 28, 36, 37, 49, 55, 56, 57, 64, 65, 71, 72, 73, 74, 77, 78, 80, 81, 82, 89, 92, 94], "api_kei": [30, 60, 61, 62], "app": [65, 93], "appar": 71, "appear": [0, 5, 6, 51, 67, 70, 82, 92, 93], "append": [27, 52, 59, 72, 82, 99], "append_paged_kv_cach": 97, "appl": 93, "appli": [0, 2, 3, 5, 7, 10, 12, 15, 16, 17, 26, 27, 28, 29, 65, 70, 71, 73, 82, 83, 87, 90, 93, 97], "applic": [9, 12, 22, 25, 26, 28, 30, 33, 34, 35, 67, 69, 71, 72, 88, 92, 93, 99], "apply_batched_logits_processor": [52, 70], "apply_chat_templ": [26, 45], "apply_llama3_sc": 82, "apply_query_key_layer_sc": [83, 84], "apply_residual_connection_post_layernorm": 84, "apply_rotary_pos_emb": 82, "apply_rotary_pos_emb_chatglm": 82, "apply_rotary_pos_emb_cogvlm": 82, "apply_silu": 82, "applybiasropeupdatekvcach": 93, "applyrop": 26, "appreci": 28, "approach": [0, 2, 4, 7, 9, 11, 12, 26, 27, 28, 69, 73, 81], "appropri": [25, 36, 92], "approv": 52, "approxim": [28, 65, 83], "apt": [20, 31, 65, 66, 67], "ar": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 33, 34, 42, 45, 46, 47, 49, 50, 52, 53, 54, 55, 56, 57, 58, 60, 61, 65, 66, 67, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 87, 88, 89, 90, 92, 93, 94, 95, 96, 97, 98, 99], "arang": 82, "arbitrag": 73, "arbitrari": [17, 93], "arbitrary_types_allow": 70, "architectur": [2, 4, 6, 9, 15, 22, 27, 28, 65, 71, 84, 87, 91, 93, 94], "arctic": [91, 93], "area": [28, 59], "aresult": 36, "arg": [0, 7, 19, 30, 54, 70, 83, 84, 87, 93], "arglist": 7, "argmax": 82, "argpars": 54, "argument": [2, 3, 20, 30, 36, 49, 52, 65, 69, 70, 73, 76, 82, 89, 93, 97], "argumentpars": 54, "aris": 65, "arithmet": 16, "armor": 51, "around": [1, 15, 19, 71, 75, 80], "arrai": [0, 1, 70, 82, 87], "arrayview": [0, 1], "arriv": [0, 4], "arrivaltim": 0, "arrow": 82, "art": [20, 26], "articl": [5, 12, 26, 27], "artifici": 71, "artist": 59, "arxiv": [0, 1, 4, 10, 29, 82, 90], "as_dtyp": 82, "as_lay": 7, "as_shap": 82, "ascii": 82, "asciichar": 1, "ask": [51, 58, 92], "aspect": 5, "assembl": [16, 18], "assert": [7, 82, 92, 93, 99], "assert_valid_quant_algo": 84, "assign": [0, 2, 19, 83, 85, 95], "assist": [6, 30, 33, 34, 45, 60, 61, 69, 88], "assistant_model": 6, "associ": [1, 3, 4, 10, 28, 65, 75, 82], "asssembl": 12, "assum": [1, 3, 9, 10, 12, 13, 20, 27, 28, 70, 73, 82, 84, 87], "assumpt": [12, 29], "async": [36, 47, 48, 70, 73, 87], "asynchron": [1, 3, 36, 40, 41, 70], "asyncio": [47, 48], "asyncllmengin": 93, "atom": 1, "attach": [2, 20], "attempt": [0, 2, 74, 75, 77], "attend": 81, "attent": [0, 1, 2, 6, 8, 9, 10, 12, 14, 16, 17, 20, 21, 29, 64, 70, 82, 87, 88, 89, 92, 93, 94, 95, 98], "attention_backend": [95, 97], "attention_head_s": [82, 83], "attention_mask": [82, 83, 84, 87, 97], "attention_mask_param": 84, "attention_mask_typ": 83, "attention_multipli": 84, "attention_output": 92, "attention_output_orig_quant_scal": 82, "attention_output_sf_scal": 82, "attention_packed_mask": [82, 83], "attention_param": [83, 84], "attention_qk_half_accumul": 93, "attention_window_s": 8, "attentionconfig": 0, "attentionheads": 1, "attentionmask": 97, "attentionmaskparam": 83, "attentionmasktyp": [82, 83], "attentionmetadata": 95, "attentionparam": [83, 84], "attentiontyp": 0, "attn_backend": [70, 97], "attn_bia": 84, "attn_dens": [10, 29], "attn_forward_funcnam": 83, "attn_k": [10, 29], "attn_logit_softcap": 84, "attn_logit_softcapping_scal": 82, "attn_metadata": 95, "attn_processor": 84, "attn_q": [10, 29], "attn_qkv": [10, 29], "attn_v": [10, 29], "attribut": [0, 1, 3, 7, 17, 19, 87], "audio": [87, 93], "audio_engine_dir": 87, "audio_featur": 87, "audio_path": 87, "authent": [69, 75, 88], "authorized_kei": [31, 32], "auto": [0, 1, 2, 3, 5, 6, 11, 13, 16, 42, 51, 70, 73, 80, 82, 84, 85, 86, 93], "auto_deploi": 93, "auto_deploy_config": 70, "auto_parallel": [29, 42, 70, 93], "auto_parallel_config": 70, "auto_parallel_world_s": [42, 70], "auto_quantize_bit": 86, "autoawq": 93, "autodeploi": 93, "autogptq": 93, "autom": [45, 93], "automat": [0, 3, 7, 11, 16, 17, 26, 30, 36, 40, 41, 69, 71, 73, 75, 82, 89, 90, 93], "autoparallelconfig": 70, "autopp": 93, "autoq": 93, "autoregress": [0, 12, 97, 98], "autotoken": 36, "autotun": [70, 93], "autotuner_en": [51, 70], "aux": 89, "auxiliari": 12, "avaiable_block": 99, "avail": [0, 1, 3, 7, 9, 16, 21, 23, 30, 36, 42, 46, 47, 49, 50, 52, 65, 71, 73, 79, 80, 81, 87, 88, 89, 90, 93, 94, 97, 98], "averag": [0, 12, 20, 27, 70, 73, 74, 75, 77, 79, 80], "avg": [73, 75, 82], "avg_pool2d": 82, "avgnumdecodedtokensperit": 0, "avgpool2d": 83, "avoid": [1, 2, 19, 26, 27, 28, 65, 69, 87, 89, 93], "awai": [79, 80], "await": [0, 3, 36, 47, 48], "awaitcontextrespons": 0, "awaitgenerationrespons": 0, "awaitrespons": [0, 2, 3], "awar": [2, 5, 21, 92], "awq": [25, 36, 59, 64, 91, 93], "awq_block_s": 86, "ax": 82, "axi": [24, 82], "b": [1, 2, 7, 10, 16, 21, 22, 23, 24, 72, 82, 84, 87, 93], "b200": [27, 28, 74, 93], "b_sf": 82, "back": [0, 2, 9, 11, 12, 46, 49, 67, 74, 93], "backbon": 71, "backend": [0, 2, 3, 12, 16, 18, 20, 27, 28, 30, 37, 45, 51, 52, 55, 56, 57, 64, 70, 72, 73, 74, 88, 93, 96, 98, 99], "backend_token": [0, 3], "backendagentdesc": 0, "backu": [0, 3, 70], "backward": 19, "bad": [0, 3, 70, 93], "bad_token_id": 70, "bad_words_data": 87, "bad_words_list": 87, "badword": 0, "badwordslen": 1, "badwordslist": 1, "badwordsptr": 1, "baichuan": [69, 90, 91, 93], "baichuan2": 91, "baichuanconfig": 84, "baichuanforcausallm": 84, "balanc": [4, 6, 12, 16, 28, 70, 79, 81], "band": 45, "bandwidth": [6, 16, 21, 22, 23, 25, 28, 45], "bangbang": 22, "bantoken": 0, "banword": 0, "bar": 70, "bare": [93, 94], "barissglc": 58, "barnardo": 51, "bart": [91, 93], "base": [0, 1, 2, 3, 9, 10, 11, 12, 14, 17, 18, 19, 20, 21, 22, 25, 26, 28, 29, 47, 48, 54, 65, 70, 71, 73, 79, 81, 82, 83, 84, 85, 86, 87, 89, 91, 93, 94, 95, 96, 98, 99], "base64": 61, "base_model": 10, "base_s": 83, "base_url": [30, 60, 61, 62], "baseagentconfig": 0, "basekvcachemanag": 0, "baselin": [25, 26, 27, 28, 75, 79, 80, 97], "baseline_fp8_engin": 77, "basellmarg": 70, "basemodel": 70, "baseresourcemanag": [96, 98], "basetransferag": 0, "bash": [16, 30, 32, 33, 34, 35, 37, 38, 39, 55, 56, 57, 72], "basic": [14, 72, 82], "basic_string_view": 0, "batch": [0, 1, 6, 9, 10, 11, 12, 13, 16, 18, 20, 22, 23, 25, 26, 27, 28, 29, 30, 52, 64, 68, 70, 73, 74, 75, 77, 78, 80, 81, 82, 83, 87, 88, 89, 92, 93, 95, 96, 97, 98, 99], "batch_beam_s": [5, 82], "batch_dim": 82, "batch_idx": 87, "batch_input_id": 87, "batch_manag": [0, 1, 98], "batch_schedul": 93, "batch_siz": [5, 7, 13, 15, 21, 24, 82, 83, 86, 87, 89, 97], "batchdon": 1, "batched_logits_processor": [52, 70], "batchedlogitsprocessor": [52, 70], "batchidx": 1, "batchindex": 1, "batching_typ": 70, "batchingtyp": [0, 70], "batchsiz": [0, 1, 6, 22], "batchsizelimit": 0, "batchsizet": 0, "batchslot": 1, "batchslotshostcopi": 1, "batchslotsrequestord": 1, "bc": 82, "beam": [0, 1, 6, 12, 18, 24, 29, 30, 36, 49, 64, 70, 82, 87, 89, 92, 93], "beam_search_diversity_r": [70, 87], "beam_width": [5, 6, 36, 82, 87, 93], "beam_width_arrai": 70, "beamhypothes": 1, "beamsearch": 0, "beamsearchbuff": 1, "beamsearchdiversityr": [0, 1, 6], "beamsiz": 0, "beamtoken": [0, 3], "beamwidth": [0, 1, 2, 3, 6, 70, 93], "beamwidtharrai": [0, 1, 6], "becam": 0, "becaus": [0, 3, 9, 20, 25, 26, 27, 28, 29, 36, 53, 58, 69, 73, 74, 75, 76, 77, 79, 81, 82, 89], "becom": [5, 6, 7, 9, 10, 16, 17, 25, 26, 28, 51, 71], "been": [0, 3, 4, 5, 19, 22, 23, 26, 28, 32, 54, 58, 65, 67, 70, 73, 77, 79, 82, 92, 93], "befor": [0, 1, 2, 3, 5, 7, 9, 10, 11, 15, 16, 17, 26, 27, 55, 56, 57, 64, 65, 67, 70, 71, 72, 76, 77, 79, 81, 82, 84, 87, 89, 92, 93, 95, 96, 97, 98, 99], "beforehand": 75, "begin": [12, 69, 71, 76, 93, 95], "behav": [0, 89], "behavior": [2, 5, 74, 79, 82, 87, 89, 93], "behaviour": [0, 82], "behind": [22, 28], "being": [0, 5, 9, 16, 19, 28, 58, 70, 79, 92, 93, 97], "believ": [51, 73], "belong": 79, "below": [0, 5, 6, 7, 8, 10, 20, 23, 24, 25, 27, 28, 31, 32, 73, 74, 77, 79, 80, 92], "bench": [20, 27, 40, 41, 58, 73, 74, 78, 93], "benchmark": [26, 27, 56, 64, 65, 72, 77, 78, 80, 88, 93], "benchmark_2nod": 30, "benefici": [28, 73, 79, 80], "benefit": [7, 9, 11, 23, 25, 27, 28, 29, 71, 79, 93], "bert": [29, 82, 90, 91, 93], "bert_attent": 82, "bert_attention_plugin": 29, "bert_context_fmha_fp32_acc": 29, "bertattent": 83, "bertattentionplugin": 82, "bertbas": 84, "bertforquestionansw": 84, "bertforsequenceclassif": [84, 91], "bertmodel": 84, "besid": 96, "best": [5, 16, 26, 27, 28, 53, 70, 72, 73, 76, 78, 79, 88, 93], "best_of": [70, 93], "best_path": 87, "best_path_len": 87, "best_path_length": 87, "best_perf_practice_on_deepseek": [26, 93], "bestpathindic": 1, "bestpathlength": 1, "beta": [30, 82], "beta_fast": 82, "beta_slow": 82, "better": [0, 2, 5, 6, 9, 11, 17, 19, 24, 26, 27, 28, 29, 55, 56, 57, 70, 74, 76, 77, 80, 81, 93], "between": [0, 2, 5, 6, 8, 9, 12, 16, 17, 19, 26, 27, 28, 34, 61, 67, 70, 72, 74, 76, 80, 81, 82, 83, 89, 92, 93, 95], "beyond": [1, 22, 77], "bf16": [5, 11, 17, 19, 26, 28, 64, 77, 80, 91, 93], "bfloat16": [5, 16, 29, 73, 75, 85, 90, 91, 93], "bhuvanesh09": 93, "bi": 5, "bia": [0, 3, 15, 16, 28, 70, 82, 83, 84, 93], "bias": [15, 82], "bidirect": [82, 83], "bidirectionalglm": 82, "bigger": 9, "biggest": 9, "billion": 20, "bin": [15, 16, 17, 20, 30, 33, 34, 35, 37, 38, 39, 55, 56, 57, 72, 92, 93], "binari": [12, 16, 72, 82], "bind": [52, 64, 70, 81, 87, 89, 93, 96, 98, 99], "bindcapacityschedul": 99, "bit": [0, 1, 5, 22, 58, 82, 90], "bitmask": 93, "bl": [12, 84], "black": 7, "blackwel": [2, 20, 27, 59, 64, 67, 76, 77, 91, 93], "blip": [90, 93], "blip2": [90, 91, 93], "blob": 26, "block": [0, 1, 2, 5, 6, 9, 16, 28, 29, 36, 51, 52, 64, 69, 70, 79, 82, 87, 89, 93, 98], "block_controlnet_hidden_st": 84, "block_hash": 51, "block_num": 82, "block_siz": [82, 83, 87], "block_sparse_block_s": 82, "block_sparse_homo_head_pattern": 82, "block_sparse_num_local_block": 82, "block_sparse_param": 83, "block_sparse_vertical_strid": 82, "blockhash": 0, "blockidx": 1, "blockptr": 1, "blocksiz": 0, "blockspars": 82, "blocksparseattnparam": 83, "blog": [20, 21, 24, 25, 26, 27, 28, 93], "bloodeagle40234": 93, "bloom": [6, 17, 90, 91, 93], "bloom_dict": 17, "bloomforcausallm": 84, "bloommodel": 84, "bm": 1, "bmm": 16, "board": 80, "bodi": 16, "book": 58, "bool": [0, 1, 7, 13, 15, 70, 82, 83, 84, 85, 87, 97], "boolean": [1, 3, 10, 82, 84, 85], "boost": [20, 26, 28, 77, 79, 80], "born": [14, 16, 92], "borrow": [36, 49, 73], "bos_token_id": 87, "both": [0, 2, 4, 5, 7, 8, 10, 12, 16, 17, 20, 22, 25, 26, 27, 28, 29, 40, 54, 70, 73, 74, 76, 79, 81, 82, 83, 89, 90, 93, 96, 97], "bottleneck": [4, 11, 20, 25, 76, 79], "bottom": 32, "bound": [0, 6, 14, 16, 23, 26, 27, 28, 70, 73, 82, 87, 89], "boundari": [6, 16, 28, 70, 82, 84, 86, 89], "box": [7, 20], "bpru": 93, "brahma": 73, "branch": [12, 21, 24, 70], "breadth": 12, "break": [12, 26, 69, 73, 80, 93, 99], "breakdown": [72, 73, 74, 75], "breviti": 20, "brief": [84, 87, 97], "briefli": [34, 61], "brife": 0, "bring": [25, 26, 27, 28, 95], "broadcast": [3, 26, 82], "broadcast_help": 82, "broader": [5, 93], "broadli": 28, "broken": [71, 79, 93], "bsz": 83, "bu": 65, "budget": [13, 79], "buffer": [0, 1, 2, 3, 8, 9, 29, 30, 64, 70, 82, 93, 98], "buffer_0": 1, "buffer_1": 1, "buffer_2": 1, "buffer_alloc": 87, "buffercast": 1, "buffercastornul": 1, "bufferdatatyp": 1, "buffermanag": 89, "buffermanagertest": 1, "bufferptr": 1, "bufferrang": 1, "buffers": 1, "bufferview": 0, "bug": [28, 93], "build": [2, 3, 5, 6, 7, 9, 10, 12, 13, 14, 16, 18, 49, 51, 53, 54, 58, 64, 69, 70, 71, 72, 76, 77, 78, 79, 81, 84, 85, 88, 89, 92, 93], "build_cach": 70, "build_config": [19, 29, 36, 49, 53, 54, 58, 70, 77, 79, 80, 84], "build_dir": 65, "build_engin": 16, "build_flags_multiple_profil": 80, "build_serialized_network": 16, "build_wheel": [20, 65, 72], "buildcacheconfig": 70, "buildconfig": [13, 19, 36, 49, 53, 54, 58, 70, 77, 79, 80, 93], "builder": [13, 16, 19, 70, 93], "builder_force_num_profil": 93, "builder_opt": 93, "built": [3, 6, 9, 16, 19, 28, 29, 59, 65, 67, 69, 70, 73, 74, 75, 80, 81, 82, 88, 89, 92, 93], "bump": 1, "bumptaskinprogress": 1, "burden": 76, "busi": 0, "button": 93, "buvnswrn": 93, "bw": 93, "byt5": [91, 93], "byte": [0, 1, 11, 70, 87], "bytestostr": 1, "c": [0, 1, 2, 5, 7, 12, 16, 18, 20, 27, 28, 30, 31, 32, 36, 55, 56, 57, 64, 70, 71, 72, 79, 82, 84, 88, 93, 96, 98, 99], "cach": [0, 1, 2, 3, 6, 10, 16, 19, 25, 26, 27, 28, 29, 30, 36, 40, 41, 43, 44, 54, 64, 68, 70, 71, 73, 74, 75, 79, 82, 87, 88, 90, 93, 94, 95, 96, 97, 99], "cache_indir": 87, "cache_indir_t": 82, "cache_indirect": [5, 82, 83, 87, 92], "cache_root": 70, "cache_transceiver_config": 70, "cachehitr": 0, "cacheindirect": 1, "cachelevel": 0, "cachelevelupd": 0, "caches": 0, "cachest": 0, "cachetransceiv": 0, "cachetransceiverconfig": [0, 70], "cachetyp": 98, "cachevalu": 1, "calcul": [0, 21, 22, 24, 27, 28, 70, 73, 81, 82, 87, 89, 93], "calculate_speculative_resourc": 70, "calculatespeculativeresourc": 0, "calculatespeculativeresourcetupl": 0, "calib_batch": [59, 70, 77, 84], "calib_batch_s": [70, 77, 84], "calib_config": [59, 70, 77], "calib_dataset": [59, 70, 84, 86], "calib_max_seq_length": [59, 70, 77, 84, 86], "calib_s": [73, 86], "calibconfig": [59, 70, 77], "calibr": [17, 25, 28, 29, 59, 70, 77, 93], "call": [0, 1, 3, 4, 5, 6, 7, 16, 17, 19, 27, 28, 36, 52, 70, 72, 75, 77, 82, 84, 86, 87, 88, 89, 93, 95, 96, 97, 98], "callabl": [17, 52, 70, 84], "callback": [3, 52, 70], "can": [0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 13, 16, 17, 18, 19, 20, 21, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 36, 40, 43, 44, 46, 49, 52, 53, 54, 55, 56, 57, 58, 59, 64, 65, 67, 69, 70, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 84, 85, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99], "canaccessp": 1, "cancel": [0, 3, 70, 73, 93], "cancelrequest": [0, 3], "candid": [0, 6, 12, 16, 26, 27, 70], "canenqueu": 0, "canenqueuerequest": 0, "cannon": 51, "cannot": [1, 6, 11, 16, 17, 26, 28, 69, 70, 79, 80, 81, 82, 89, 92, 93, 99], "cap": 75, "capabl": [21, 26, 45, 65, 71, 72, 77], "capac": [0, 1, 21, 23, 25, 70, 99], "capacitor_schedul": 99, "capacity_scheduler_polici": [70, 81], "capacityschedul": [96, 98, 99], "capacityschedulerpolici": [0, 70, 81, 93], "capit": [40, 42, 43, 44, 46, 47, 48, 49, 50, 54, 59, 66, 67, 75, 81, 88, 94], "caption": 83, "captur": [27, 28, 70, 97], "card": [53, 58], "carefulli": 20, "case": [0, 1, 2, 5, 6, 8, 9, 10, 12, 20, 22, 25, 26, 27, 28, 29, 36, 73, 74, 75, 77, 78, 80, 82, 90, 93], "cast": [28, 82], "cast_to_dtyp": 82, "castsiz": 1, "cat": [20, 27, 30, 56], "categor": [12, 28, 82], "categori": 85, "categorical_sampl": 82, "caus": [2, 3, 17, 19, 29, 70, 80, 92, 93], "causal": [27, 82, 83, 97], "cautiou": 19, "caveat": 77, "cd": [14, 15, 20, 27, 65, 73, 88, 92, 94], "ceil": [1, 84], "ceil_mod": [82, 83], "ceildiv": 1, "center": [22, 23], "central": 85, "certain": [2, 7, 15, 67, 71, 82], "cg": 84, "chain": 27, "challeng": [26, 71], "chanc": [9, 29, 81], "chang": [2, 5, 6, 8, 9, 10, 17, 19, 21, 23, 24, 27, 28, 65, 69, 70, 71, 73, 80, 82, 84, 87, 89, 92, 94, 98], "channel": [29, 82, 90, 93], "char": [0, 1], "charg": [6, 16, 97], "chart": 22, "chat": [12, 23, 35, 38, 40, 42, 45, 46, 47, 48, 49, 50, 51, 52, 53, 58, 59, 62, 63, 66, 67, 69, 88, 93, 94], "chatbot": 58, "chatcmpl": 88, "chatglm": [69, 82, 90, 91, 93], "chatglm2": [69, 91, 93], "chatglm3": [69, 84, 91, 93], "chatglm_vers": 84, "chatglmconfig": 84, "chatglmforcausallm": 84, "chatglmgenerationsess": 87, "chatglmmodel": 84, "check": [2, 3, 40, 66, 67, 70, 74, 76, 77, 79, 80, 82, 87, 88, 89, 92, 93, 95], "check_accuraci": 15, "check_config": 84, "check_gpt_mem_usag": 89, "checkbeamsearchdiversityr": 0, "checkbeamwidth": 0, "checkbeamwidtharrai": 0, "checkearlystop": 0, "checklengthpenalti": 0, "checkminp": 0, "checkmintoken": 0, "checknorepeatngrams": 0, "checknumreturnsequ": 0, "checkpoint": [14, 17, 18, 19, 20, 26, 27, 28, 29, 30, 46, 54, 64, 69, 70, 73, 75, 77, 86, 87, 88, 90, 92, 93, 95], "checkpoint_dir": [10, 13, 14, 15, 16, 19, 29, 73, 88, 92], "checkposteriorvalu": 0, "checkremotedesc": 0, "checkrepetitionpenalti": 0, "checktemperatur": 0, "checktopk": 0, "checktopp": 0, "checktoppdecai": 0, "checktoppmin": 0, "checktoppresetid": 0, "chef": 92, "chmod": 31, "choic": [0, 12, 25, 27, 29, 54, 73, 76, 82, 87, 88, 97], "choos": [16, 19, 26, 28, 77, 82, 93], "chosen": [28, 89, 99], "chrome": 72, "chrono": 0, "chunk": [0, 8, 28, 29, 64, 68, 70, 80, 82, 87, 89, 93], "chunk_dim": 83, "chunk_length": 93, "chunk_scan": 82, "chunk_siz": [82, 84], "chunkedcontextnexttoken": 1, "chunkedcontextnexttokenshost": 1, "ci": 1, "circular": 5, "citi": [59, 88], "ckpt": [54, 73, 88], "ckpt_dir": [16, 19, 84], "ckpt_llama_3": 16, "cl": [14, 19], "claim": [1, 17], "claimpag": 1, "claimpageswithevict": 1, "clamp": [70, 93], "clamp_val": 70, "class": [0, 1, 2, 5, 6, 7, 8, 13, 14, 16, 17, 19, 25, 29, 36, 43, 44, 46, 49, 52, 53, 54, 65, 69, 70, 76, 77, 80, 82, 83, 84, 85, 86, 87, 92, 93, 95, 96, 97, 99], "class_dropout_prob": 83, "class_label": 83, "classic": [16, 64], "classifi": [83, 84], "classmethod": [14, 19, 70, 83, 84, 87], "classvar": 70, "clean": [20, 65, 72, 92], "clear": [67, 79, 87], "clearli": 81, "cli": [15, 20, 36, 64, 73, 76, 77, 79, 80, 88], "click": [31, 32], "client": [0, 3, 30, 63, 74], "client_id": 52, "clientid": 0, "clip": 82, "clip_before_cast": 82, "clip_qkv": [83, 84], "clip_vision_model": 84, "clipvisiontransform": 84, "clock": 26, "clone": [10, 20, 65, 69, 75, 88, 92, 94], "clone_input": 7, "close": [5, 19, 20, 29, 80, 89], "closur": 82, "cloud": [22, 31, 32], "cls_token": 83, "cluster": [6, 16, 26, 29, 30, 67, 70, 93], "cluster_info": 93, "cluster_kei": [29, 93], "cluster_s": 30, "cmake": [65, 93], "cnn_dailymail": [59, 70, 84], "co": [0, 10, 20, 27, 28, 34, 61, 69, 82, 83, 88, 92], "coalesc": 52, "coast": 88, "code": [2, 5, 7, 8, 11, 12, 16, 19, 25, 26, 28, 30, 36, 55, 56, 57, 64, 69, 70, 71, 72, 73, 82, 90, 91, 92, 93, 95, 98, 99], "codebas": [8, 95], "codellama": 93, "codepath": 93, "codeqwen": 93, "coderham": 93, "cogvlm": [91, 93], "cogvlmattent": 83, "cogvlmconfig": 84, "cogvlmforcausallm": 84, "coher": [6, 93], "cohereconfig": 84, "cohereforcausallm": 84, "collabor": [6, 26, 28, 59, 82], "collect": [1, 7, 11, 12, 16, 26, 28, 70, 74, 82, 95], "collect_and_bia": 83, "color": [58, 79], "column": [10, 82, 90], "columnlinear": [10, 14, 83], "com": [19, 20, 26, 65, 82, 88, 92, 93, 94], "combin": [0, 7, 12, 23, 26, 27, 28, 29, 54, 55, 56, 57, 73, 74, 77, 79, 83, 93, 97, 99], "combinedtimesteplabelembed": 83, "combinedtimesteptextprojembed": 83, "come": [6, 10, 22, 75, 76, 79, 81, 89, 92], "comm": 70, "comma": [82, 87], "command": [9, 10, 14, 15, 16, 19, 20, 30, 31, 32, 55, 56, 57, 65, 69, 72, 73, 75, 80, 85, 88, 89, 92, 93, 94], "commandr": 93, "comment": 93, "commit": 28, "commmod": 0, "common": [0, 5, 8, 9, 12, 20, 28, 40, 51, 69, 70, 82, 89, 98], "common_prefix": 51, "commonli": [7, 26, 30, 93], "commstat": 0, "commtyp": 0, "commun": [0, 2, 6, 11, 16, 28, 29, 59, 69, 71, 77, 82, 91, 93], "communicationmod": [0, 2], "communicationtyp": 0, "compani": 53, "compar": [1, 2, 17, 22, 23, 25, 27, 28, 77, 79, 80, 81, 82, 97], "comparison": [6, 22, 26, 27, 73], "compat": [12, 19, 27, 30, 65, 80, 83, 88, 91, 93, 95], "compbin": 10, "compil": [6, 11, 18, 64, 67, 70, 71, 72, 73, 82, 92], "complet": [0, 1, 2, 3, 6, 8, 9, 12, 33, 34, 36, 60, 61, 63, 65, 69, 70, 71, 73, 74, 75, 79, 80, 88, 93, 98, 99], "completion_token": 88, "completionoutput": [36, 53, 70], "complex": [7, 8, 12, 16, 26], "compli": 30, "complic": [27, 28, 95], "compon": [2, 3, 5, 16, 18, 25, 26, 27, 28, 64, 90, 96], "compos": [0, 6, 73], "comprehens": [20, 30, 71], "compress": [21, 28], "compris": 25, "comput": [0, 1, 4, 5, 6, 9, 12, 16, 21, 22, 23, 25, 26, 27, 28, 29, 42, 46, 47, 49, 50, 52, 70, 72, 73, 76, 77, 81, 82, 89, 92, 93, 95, 96, 97, 98], "compute_relative_bia": 83, "computecontextlogit": 1, "computegenerationlogit": 1, "computenumpackedmask": 1, "concat": [14, 26, 82], "concat_kvcach": 26, "concaten": [5, 10, 17, 26, 82, 95], "conced": 51, "concept": [16, 73, 78, 93, 98], "conceptu": 1, "concern": [16, 89], "conclus": 78, "concret": 95, "concur": 51, "concurr": [1, 2, 12, 20, 22, 26, 27, 28, 73, 93], "cond_proj_dim": 83, "conda": 93, "condit": [0, 1, 3, 6, 7, 12, 73, 82, 83, 93], "condition": 82, "conditioning_embed": 83, "conditioning_embedding_dim": 83, "conduct": [5, 73], "confess": 51, "config": [0, 1, 5, 9, 10, 13, 14, 17, 19, 20, 21, 27, 28, 30, 37, 70, 73, 79, 83, 84, 85, 87, 92, 93, 95, 98], "config_class": 84, "config_dir": 84, "config_fil": [30, 70, 84], "configdict": 70, "configur": [0, 1, 2, 4, 5, 8, 12, 17, 18, 20, 23, 29, 30, 43, 44, 45, 49, 53, 54, 58, 65, 67, 70, 73, 74, 75, 78, 79, 81, 84, 87, 89, 92, 93, 97], "configuration_llama": 95, "configuration_mymodel": 95, "configuration_util": 95, "confirm": [42, 46, 47, 49, 50], "conform": 70, "conjunct": 79, "connect": [0, 11, 16, 75, 76, 78], "connectioninfo": 0, "connectioninfotyp": 0, "connectionmanag": 0, "connectremoteag": 0, "consecut": 6, "consequ": [2, 25, 76, 80], "conserv": [0, 81], "consid": [0, 1, 10, 12, 20, 25, 58, 59, 70, 74, 79, 82, 95, 99], "consider": [19, 25, 36], "consist": [7, 19, 22, 26, 70, 71, 73, 75, 82, 90, 92, 97], "consol": 31, "consolid": 12, "const": [0, 1, 3], "const_iter": 1, "constant": [1, 5, 82, 89], "constant_to_tensor_": 82, "constantli": [42, 46, 47, 49, 50], "constants_to_tensors_": 82, "constantthreshold": 1, "constexpr": [0, 1], "constpointercast": 1, "constrain": [6, 25], "constraint": [0, 5, 6, 25, 67, 82], "construct": [0, 1, 3, 12, 16, 73, 82, 93, 97], "constructor": [0, 13, 58, 69, 88, 97], "consult": [12, 65, 72], "consum": [0, 7, 28, 70, 82], "consumpt": [5, 22, 27, 29], "contact": 82, "contain": [0, 1, 2, 3, 5, 6, 7, 8, 10, 11, 15, 16, 17, 18, 19, 26, 29, 30, 32, 55, 56, 57, 66, 67, 70, 71, 73, 74, 82, 84, 87, 88, 90, 91, 93, 94, 96, 97], "container_imag": [55, 56, 57], "container_img": 30, "content": [1, 10, 19, 30, 31, 33, 34, 35, 45, 60, 61, 64, 82, 88, 89, 93], "context": [0, 2, 4, 9, 25, 27, 28, 29, 64, 68, 70, 73, 78, 82, 87, 89, 92, 93, 97, 98, 99], "context_chunking_polici": [70, 81], "context_fmha": [10, 29], "context_fmha_fp32_acc": 93, "context_fmha_typ": [5, 89], "context_init": 99, "context_len": [87, 97], "context_length": [82, 83, 87, 92], "context_logit": [70, 87], "context_mem_s": 87, "context_onli": 70, "context_parallel_s": 70, "context_phas": 5, "context_pre_onli": 83, "context_request": 99, "contextchunkingpolici": [0, 70, 81, 93], "contextexecutor": 2, "contextfmha": 1, "contextidx": 0, "contextlogit": 0, "contextmanag": 69, "contextparallel": 1, "contextphaseparam": [0, 2, 70], "contextpositionid": 1, "contextprefillposit": 0, "contextrequest": 1, "contextrequestid": 2, "contextrespons": 2, "contigu": [2, 8, 76, 82, 93], "continu": [1, 3, 5, 12, 23, 25, 29, 70, 71, 77, 79, 87, 99], "contract": 73, "contrast": [6, 12, 97], "contrib": 21, "contribut": [19, 27, 28, 73, 82, 93], "contributor": [26, 89], "control": [0, 2, 5, 6, 7, 11, 36, 40, 41, 70, 72, 73, 75, 81, 82, 83, 87, 90, 93], "conv": 82, "conv1d": [29, 82, 83], "conv2d": [82, 83], "conv3d": [82, 83], "conv_bia": 82, "conv_kernel": 87, "conv_stat": 84, "conv_state_or_ptr": 82, "conv_transpose2d": 82, "conv_weight": 82, "conveni": [1, 14, 19, 65], "convent": [19, 82], "convers": [1, 17, 24, 25, 58, 64, 88, 93], "convert": [0, 1, 10, 13, 14, 15, 16, 17, 19, 71, 73, 75, 77, 88, 92, 93, 97], "convert_and_load_weights_into_trtllm_llama": 19, "convert_checkpoint": [10, 13, 14, 15, 16, 19, 75, 76, 88, 92, 93], "convert_coneckpoint": 4, "convert_hf_mpt_legaci": 93, "convert_load_format": 70, "convert_util": 93, "convert_weights_from_custom_training_checkpoint": 19, "convkernel": 1, "convolut": [0, 87], "convtranspose2d": 83, "coordin": [12, 64, 82], "copi": [0, 1, 2, 9, 12, 29, 32, 70, 77, 82, 89, 93, 97], "copy_on_partial_reus": 70, "copyfrom": 1, "copyonpartialreus": 0, "copytask": 1, "copytaskmappag": 1, "copyto": 0, "copytocpu": 0, "copytogpu": 0, "copytomanag": 0, "copytopag": 1, "copytopin": 0, "copytopooledpin": 0, "core": [6, 7, 10, 13, 16, 19, 21, 22, 24, 28, 65, 69, 73, 76, 88, 92, 93, 96], "corner": 28, "coroutin": [47, 48, 70], "correct": [2, 3, 5, 10, 12, 27, 93], "correctli": [9, 82, 93, 95], "correspond": [0, 1, 2, 4, 5, 7, 8, 10, 12, 17, 19, 27, 30, 70, 72, 80, 82, 83, 87, 90, 92, 93, 95], "cost": [9, 16, 26, 27, 28, 73, 76, 89, 93], "costli": 26, "could": [0, 2, 7, 8, 9, 15, 46, 47, 48, 49, 50, 59, 70, 75, 89, 92, 93], "couldn": 79, "count": [0, 1, 6, 30, 38, 39, 69, 73, 84, 88], "count_include_pad": [82, 83], "countlocallay": 1, "countlowerranklay": 1, "cours": 12, "court": [42, 46, 47, 49, 50], "cover": [20, 77, 78, 80], "coverag": 70, "cp312": 65, "cp_config": 70, "cp_group": [82, 83], "cp_rank": [82, 83], "cp_size": [82, 83, 86, 93], "cp_split_plugin": 82, "cpp": [2, 3, 5, 6, 16, 20, 28, 30, 56, 64, 65, 72, 73, 74, 75, 92, 93], "cpp_e2e": 87, "cpp_extens": 67, "cpp_llm_onli": 87, "cpp_onli": 65, "cpu": [0, 1, 8, 9, 10, 13, 16, 26, 27, 29, 30, 52, 67, 70, 82, 89, 92, 93, 97], "cpumemusag": [0, 70], "crash": 93, "creat": [1, 2, 3, 7, 8, 9, 12, 13, 14, 16, 18, 19, 26, 30, 31, 36, 42, 46, 47, 48, 49, 50, 51, 52, 59, 60, 61, 62, 69, 70, 71, 73, 74, 75, 79, 80, 82, 83, 84, 87, 88, 89, 93, 95, 96, 97, 99], "create_allreduce_plugin": 82, "create_attention_const_param": 83, "create_builder_config": 13, "create_cuda_graph_metadata": 97, "create_execution_context": 87, "create_fake_weight": 82, "create_network": 16, "create_pytorch_model_based_executor": [98, 99], "create_runtime_default": 84, "create_sinusoidal_posit": 82, "create_sinusoidal_positions_for_attention_plugin": 82, "create_sinusoidal_positions_for_cogvlm_attention_plugin": 82, "create_sinusoidal_positions_long_rop": 82, "create_sinusoidal_positions_yarn": 82, "createloramodul": 1, "creation": [1, 70, 82, 89], "creativ": 6, "criteria": 87, "critic": [26, 73, 92], "crop": 83, "cropped_pos_emb": 83, "cross": [0, 10, 11, 26, 27, 70, 82, 87, 93], "cross_attent": [83, 87], "cross_attention_dim": 83, "cross_attention_mask": [83, 87], "cross_attention_mask_for_context": 87, "cross_attention_mask_for_gen": 87, "cross_attention_norm": 83, "cross_attention_norm_num_group": 83, "cross_attention_packed_mask": 83, "cross_attn_dens": [10, 29], "cross_attn_k": [10, 29], "cross_attn_q": [10, 29], "cross_attn_qkv": [10, 29], "cross_attn_v": [10, 29], "cross_kv": 82, "cross_kv_cache_block_offset": [83, 87], "cross_kv_cache_fract": [70, 87], "cross_kv_cache_gen": [83, 84], "cross_kv_length": 82, "cross_kv_reus": [83, 84], "crossattentionmask": 0, "crosskvcachefract": [0, 93], "crosskvcachestat": 0, "crucial": [12, 16, 25, 96], "ctor": 82, "ctx": 0, "ctx_request_id": 70, "ctxenginepath": 0, "ctxexecutorconfig": 0, "cu": [16, 26], "cu12": 93, "cu128": [66, 67], "cuassert": 92, "cubla": 28, "cublaslt": [29, 80], "cublasltmatmul": 28, "cublasscaledmm": 28, "cuda": [0, 1, 2, 5, 11, 16, 20, 27, 28, 52, 59, 65, 66, 67, 70, 72, 73, 84, 87, 89, 92, 93, 97, 98], "cuda_arch": 65, "cuda_architectur": [20, 65], "cuda_graph_batch_s": [20, 70, 74], "cuda_graph_cache_s": 70, "cuda_graph_inst": 92, "cuda_graph_max_batch_s": 70, "cuda_graph_mod": [70, 87, 92], "cuda_graph_padding_en": [20, 28, 56, 70, 74], "cuda_hom": 67, "cuda_launch_block": 92, "cuda_stream": 92, "cuda_stream_guard": 87, "cuda_stream_sync": 82, "cudadevicegetstreampriorityrang": 1, "cudaevent_t": 1, "cudaeventdisabletim": 1, "cudagraph": 93, "cudagraphcaches": 0, "cudagraphlaunch": 92, "cudagraphmod": 0, "cudamalloc": [1, 2], "cudamallocasync": [1, 2], "cudamemcpyasync": 52, "cudamempool": 1, "cudamempoolptr": 1, "cudaprofilerapi": 72, "cudart": 92, "cudastream": 0, "cudastream_t": 1, "cudastreamcreatewithflag": 1, "cudastreamnonblock": 1, "cudastreamptr": [0, 1], "cudeviceptr": 1, "cudnn": 93, "cufil": 0, "cumemgenericallocationhandl": 1, "cumlogprob": [0, 1], "cumlogprobscba": 1, "cumsum": [82, 93], "cumsumgenerationlength": 1, "cumsumlastdim": 82, "cumsumlength": 1, "cumul": [0, 1, 70, 82], "cumulative_logprob": [36, 53, 70], "curand": 93, "curl": [30, 63, 88], "currenc": 73, "current": [0, 1, 2, 3, 5, 10, 12, 20, 25, 26, 27, 28, 29, 36, 45, 58, 65, 70, 73, 77, 79, 80, 81, 82, 87, 89, 91, 93, 94, 96, 97, 98, 99], "current_stream": 92, "currentexpandindic": 1, "curv": 24, "custom": [6, 16, 19, 21, 26, 27, 29, 40, 41, 43, 44, 52, 53, 54, 65, 71, 77, 80, 82, 87, 93, 96, 97], "custom_all_reduc": 93, "custom_mask": 82, "customallreduc": 93, "customized_key_dict": 17, "customized_preprocess": 17, "customizedmodulea": 17, "customizedmoduleb": 17, "cutlass": [28, 70, 93], "cxx11": 65, "cyclic": [64, 82, 87], "d": [1, 10, 30, 31, 33, 34, 35, 55, 56, 57, 58, 73, 82, 83, 88, 92, 93], "d0": 26, "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b": 73, "d_": 27, "d_6": 27, "dangl": 7, "data": [0, 1, 2, 5, 6, 8, 11, 16, 17, 21, 22, 23, 24, 25, 26, 29, 51, 61, 70, 73, 74, 75, 82, 84, 91, 92, 93, 95], "data_path": 56, "data_typ": [13, 15], "datacontext": 0, "dataset": [26, 27, 28, 34, 56, 59, 61, 70, 72, 77, 93], "dataset_fil": 74, "dataset_path": 73, "datatyp": [0, 1, 6, 16, 82, 87, 90, 92], "datatypetrait": 1, "date": 19, "datetim": 70, "dbrx": [90, 91, 93], "dbrxconfig": 84, "dbrxforcausallm": 84, "dconv": 82, "de": 1, "deactiv": 36, "dead": 93, "deal": [5, 7, 92], "dealloc": [1, 8, 99], "death": [42, 46, 47, 49, 50], "debug": [0, 8, 29, 30, 64, 65, 87, 89, 93], "debug_buff": 92, "debug_mod": [87, 92], "debug_tensors_to_sav": 87, "debugconfig": 0, "debuginputtensor": 0, "debugoutputtensor": 0, "debugtensor": 0, "debugtensornam": 0, "debugtensorsmaxiter": 0, "debugtensorsperiter": 0, "dec": [29, 87, 93], "decai": [0, 6, 70], "decid": [5, 15, 64, 73, 78, 79, 90, 96, 99], "decilmforcausallm": 91, "decis": [58, 82], "declar": [1, 6, 7, 19, 96, 98], "decltyp": [0, 1], "decod": [0, 1, 2, 5, 6, 14, 19, 26, 28, 30, 40, 41, 64, 70, 73, 82, 87, 91, 93, 95, 98], "decode_batch": 87, "decode_duration_m": 70, "decode_regular": 87, "decode_retention_prior": 70, "decode_stream": 87, "decode_words_list": 87, "decode_wrapp": 97, "decodedurationm": 0, "decoder_batch": 1, "decoder_input_id": [84, 87], "decoder_language_adapter_rout": 87, "decoder_lay": 95, "decoder_start_token_id": 29, "decoderbuff": 1, "decoderenginebuff": 0, "decoderetentionprior": 0, "decoderjsonconfigstr": 0, "decoderlay": 95, "decoderlayerlist": 14, "decoderlookaheadbuff": 1, "decodermaskedmultiheadattent": 5, "decodermodel": [0, 84, 95], "decodermodelforcausallm": [14, 19, 84, 95], "decodermodelpath": 0, "decoderst": 93, "decoderxqarunn": 5, "decoding_config": 70, "decoding_typ": [20, 27, 70], "decodingbaseconfig": 70, "decodingconfig": [0, 1], "decodinginputptr": 1, "decodingit": 0, "decodinglayerworkspac": 1, "decodingmod": [0, 1, 93], "decodingoutputptr": 1, "decompos": 5, "decor": 95, "decoupl": [26, 89], "decreas": [21, 22, 77], "dedic": [26, 28, 92], "deduc": [29, 30, 93], "deep": [16, 22, 23, 72, 82, 93], "deeper": 27, "deepgemm": 20, "deeplearn": [82, 92], "deepseek": [30, 63, 72, 74, 91, 93], "deepseek_v1": 93, "deepseek_v2": 93, "deepseek_v3": [26, 93], "deepseekforcausallm": 84, "deepseekv1config": 84, "deepseekv2": 82, "deepseekv2attent": 83, "deepseekv2config": 84, "deepseekv2forcausallm": 84, "deepseekv3forcausallm": 91, "deepseekv3routingimpl": 28, "deepspe": 15, "def": [7, 14, 16, 17, 19, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 66, 67, 75, 77, 80, 81, 88, 92, 94, 95, 99], "default": [0, 1, 2, 3, 4, 5, 6, 9, 15, 17, 19, 27, 28, 29, 30, 31, 36, 54, 59, 64, 65, 70, 72, 74, 77, 78, 79, 80, 81, 82, 84, 87, 88, 89, 90, 92, 93, 95, 97], "default_net": 82, "default_plugin_config": 84, "default_trtnet": 16, "defaultvalu": 1, "defer": 82, "defin": [0, 1, 3, 5, 7, 12, 15, 16, 17, 18, 19, 20, 23, 29, 71, 73, 80, 82, 83, 90, 93, 95, 97], "definit": [3, 5, 8, 18, 19, 26, 64, 71, 82, 92], "deftruth": 93, "degrad": [0, 2, 29, 77], "degre": [42, 46, 47, 49, 50, 52, 74, 77, 80], "delai": [74, 93], "deleg": [82, 97], "delet": [0, 1, 85, 92], "deliv": [20, 21, 24, 26, 27, 74], "delta": [0, 26, 27, 82, 83], "delta_bia": 82, "delta_softplu": 82, "delv": 28, "demand": [26, 28], "demo": [26, 34, 61], "demonstr": [3, 17, 22, 26, 69, 75, 77, 79, 80], "denmark": 51, "denois": 83, "denot": 12, "dens": [4, 5, 10, 15, 17, 82], "dense_4h_to_h": 17, "dense_bia": 83, "dense_h_to_4h": 17, "densiti": 25, "dep": 65, "departur": 51, "depend": [0, 2, 3, 5, 6, 7, 12, 15, 23, 30, 67, 70, 74, 75, 77, 80, 82, 89, 92, 93, 98], "deploi": [12, 15, 30, 64, 67, 70, 71], "deplot": [91, 93], "deploy": [25, 26, 71, 73, 77, 88, 93], "deprec": [29, 70, 71, 73, 93], "deprecationwarn": 73, "depriv": 7, "depth": 12, "dequ": [0, 1], "dequant": [5, 11, 64, 82], "deregistermemori": 0, "deriv": [16, 17, 82, 89, 96], "desc": 0, "descendli": 6, "describ": [0, 5, 6, 8, 9, 10, 12, 14, 16, 17, 18, 20, 24, 32, 34, 61, 65, 69, 73, 74, 80, 82, 90, 92, 97], "descript": [0, 1, 6, 10, 30, 54, 64, 73, 74, 80, 82, 97], "descriptor": 70, "deseri": [0, 19, 52], "deserializeadditionalmodeloutput": 0, "deserializeadditionaloutput": 0, "deserializeagentst": 0, "deserializebool": 0, "deserializecachest": 0, "deserializecachetransceiverconfig": 0, "deserializecommst": 0, "deserializecontextphaseparam": 0, "deserializedatatransceiverst": 0, "deserializedebugconfig": 0, "deserializedecodingconfig": 0, "deserializedecodingmod": 0, "deserializedisservingrequeststat": 0, "deserializedynamicbatchconfig": 0, "deserializeeagleconfig": 0, "deserializeexecutorconfig": 0, "deserializeextendedruntimeperfknobconfig": 0, "deserializeexternaldrafttokensconfig": 0, "deserializeguideddecodingconfig": 0, "deserializeguideddecodingparam": 0, "deserializeinflightbatchingstat": 0, "deserializeiterationstat": 0, "deserializeiterationstatsvec": 0, "deserializekvcacheconfig": 0, "deserializekvcacheretentionconfig": 0, "deserializekvcachestat": 0, "deserializelookaheaddecodingconfig": 0, "deserializeloraconfig": 0, "deserializemodeltyp": 0, "deserializemropeconfig": 0, "deserializeorchestratorconfig": 0, "deserializeoutputconfig": 0, "deserializeparallelconfig": 0, "deserializepeftcacheconfig": 0, "deserializeprompttuningconfig": 0, "deserializerequest": 0, "deserializerequestperfmetr": 0, "deserializerequeststag": 0, "deserializerequeststat": 0, "deserializerequeststatsperiter": 0, "deserializerequeststatsperiterationvec": 0, "deserializerespons": 0, "deserializeresult": 0, "deserializesamplingconfig": 0, "deserializeschedulerconfig": 0, "deserializesocketst": 0, "deserializespecdecfastlogitsinfo": 0, "deserializespecdecodingstat": 0, "deserializespeculativedecodingconfig": 0, "deserializestaticbatchingstat": 0, "deserializestr": 0, "deserializetensor": 0, "deserializetimepoint": 0, "deserializetokenrangeretentionconfig": 0, "design": [1, 11, 12, 16, 17, 19, 20, 25, 26, 27, 28, 69, 75, 88, 96, 97, 98], "desir": [3, 74, 82, 88, 97], "destin": [55, 56, 57], "destroi": [1, 89], "destroyipcmemori": 1, "destructor": 1, "detail": [0, 3, 5, 11, 12, 14, 16, 20, 26, 28, 29, 30, 36, 40, 45, 59, 64, 73, 74, 75, 77, 81, 82, 84, 89, 92, 93, 96, 97, 98], "detect": [0, 3, 30, 70, 82, 93], "detect_format": 17, "determin": [0, 1, 5, 6, 10, 11, 19, 27, 70, 76, 77, 81, 82, 84, 90, 96, 98, 99], "determinenumpag": 1, "determinist": [27, 80, 93], "detoken": [70, 93, 96], "detokenizedgenerationresultbas": 70, "dev": [66, 67, 93], "devel": [31, 32, 65], "develop": [14, 15, 16, 19, 26, 27, 31, 42, 46, 47, 49, 50, 64, 65, 69, 71, 75, 82, 91, 93, 95], "deviat": 74, "devic": [0, 1, 2, 52, 70, 77, 82, 84, 86, 87, 92], "device_id": 87, "device_map": 86, "device_memory_size_v2": 89, "device_request_typ": 84, "deviceallocationnvl": 1, "devicecach": 1, "devicecacheperc": 0, "deviceid": [0, 1, 2], "dgx": [6, 16, 20, 28], "di": 27, "diagon": 82, "diagram": [12, 28], "diamond": [26, 28], "dict": [14, 17, 19, 70, 82, 84, 87, 93, 95, 98], "dict_kei": 92, "dictat": 79, "dictionari": [15, 17, 70, 83], "didn": 79, "differ": [0, 1, 2, 4, 5, 6, 8, 9, 11, 14, 15, 16, 17, 19, 20, 25, 27, 28, 29, 34, 61, 65, 69, 70, 71, 73, 75, 77, 79, 80, 82, 84, 87, 89, 90, 93, 97], "differenti": 82, "difftyp": 1, "diffus": [34, 61, 83, 93], "diffusersattent": 83, "digit": 71, "dilat": [82, 83], "dim": [0, 1, 82, 83, 84, 87, 92], "dim0": 82, "dim1": 82, "dim_head": 83, "dim_in": 83, "dim_out": 83, "dim_rang": 82, "dimems": 1, "dimens": [0, 1, 5, 6, 10, 28, 82, 83, 84, 89, 92, 93, 95], "dimension": 82, "dimrang": 82, "dimtype64": [0, 1], "dir": [36, 65, 69], "direct": [0, 2, 11, 19, 67, 92], "directli": [0, 2, 6, 7, 12, 16, 19, 27, 28, 32, 36, 65, 69, 73, 80, 81, 82, 88, 93, 97, 99], "directori": [0, 3, 14, 15, 16, 17, 19, 29, 55, 56, 57, 65, 70, 73, 74, 75, 84, 87, 88, 93, 95], "disabl": [0, 1, 5, 6, 9, 13, 17, 29, 70, 73, 77, 80, 81, 82, 85, 87, 89, 93], "disable_forward_chunk": 84, "disable_kv_cach": 87, "disable_overlap_schedul": [28, 70], "disable_weight_only_quant_plugin": 84, "disable_xqa": 5, "disablelookahead": 1, "disablelookaheaddecod": 1, "disableseamlesslookaheaddecod": 1, "disadvantag": [19, 76], "disagg_executor": 0, "disaggexecutororchestr": [0, 2], "disaggreg": [0, 64, 70, 93], "disaggregated_param": 70, "disaggregatedparam": 70, "disaggserverbenchmark": [2, 93], "disaggserverutil": 2, "discard": 77, "disclaim": [27, 75, 77, 79, 80], "disclosur": 93, "disconnect": 93, "discourag": [0, 6, 70], "discov": [16, 67], "discrep": [65, 95], "discuss": [5, 27, 75, 77, 80, 81, 93], "disk": [3, 19, 46, 49, 65, 69], "dispatch": [0, 4, 19, 26, 36], "displai": 70, "disservingrequeststat": 0, "disservingstat": 0, "dist": [20, 56, 67, 72, 73, 74, 75], "distanc": [5, 82], "distil": 93, "distinct": [8, 10, 12, 26, 82], "distinguish": 9, "distribut": [1, 4, 5, 6, 16, 26, 40, 41, 73, 82, 87, 89], "distserv": 2, "disturb": 51, "dit": [84, 93], "div": 82, "dive": [27, 71, 72], "divers": [0, 6, 72], "diversity_penalti": 6, "divid": [17, 27, 82, 93], "divup": 82, "dl": 25, "dlsym": 0, "do": [1, 2, 7, 17, 19, 20, 25, 26, 27, 28, 36, 64, 67, 70, 75, 77, 80, 82, 88, 92, 95, 97], "do_cross_attent": [82, 83], "do_layer_norm_befor": 15, "do_sampl": 6, "doc": [1, 20, 24, 26, 32, 77, 80, 82, 92, 93], "docker": [20, 55, 56, 57, 64, 88, 92, 93], "docker_run_arg": 20, "dockerfil": [31, 65], "document": [0, 2, 5, 6, 8, 9, 10, 12, 14, 15, 16, 18, 19, 22, 23, 25, 27, 33, 34, 35, 36, 37, 38, 39, 45, 60, 61, 62, 65, 67, 68, 72, 74, 75, 81, 82, 89, 90, 92, 96, 97], "doe": [0, 2, 5, 10, 12, 19, 20, 21, 28, 29, 73, 74, 80, 82, 87, 89, 91, 93, 95, 99], "doesn": [1, 5, 26, 31, 36, 73, 79, 80], "dollar": 73, "domain": 11, "domin": [26, 93], "don": [12, 19, 28, 31, 76, 80, 82], "done": [1, 9, 16, 20, 28, 71, 73, 77, 79, 82, 85, 95], "dongjiyingdji": 93, "dora": [29, 82, 83], "dora_plugin": [10, 29, 82], "dot": [17, 26, 82], "doubl": [0, 22, 78, 80, 92], "down": [0, 2, 3, 10, 21, 27, 28, 58, 71, 76, 82, 87], "down_proj": 17, "download": [18, 55, 56, 57, 58, 65, 66, 67, 69, 73, 75, 88, 92, 93], "downscale_freq_shift": 83, "downsid": 80, "downstream": 90, "dp": [20, 21, 24, 26, 27, 28, 93], "dp8": [26, 28], "dprank": 0, "dpsize": 0, "dq": 64, "draft": [0, 1, 26, 27, 29, 64, 70, 87, 93], "draft_indic": 84, "draft_len": 84, "draft_path": 87, "draft_prob": 84, "draft_target_model": 12, "draft_token": [70, 84], "draft_tokens_extern": [29, 84], "draftacceptancethreshold": 1, "draftbuff": 1, "drafter": [12, 70], "draftindic": 1, "draftlen": 1, "draftlogit": 1, "draftoverhead": 0, "draftparticipantid": 0, "draftpath": 1, "draftpathshost": 1, "draftprob": 1, "draftrequestid": 0, "drafttoken": [0, 1], "drafttokenid": 1, "drafttokensextern": 1, "dram": [0, 16, 70], "drastic": 28, "dreamgenx": 93, "drive": [16, 73], "driven": 71, "driver": [89, 93], "drop": [27, 28, 77, 79, 81], "dropout": 83, "dropout_prob": 83, "dry_run": [29, 70, 93], "dst": 1, "dstate": 82, "dstdesc": 0, "dsttype": 1, "dt_proj": 82, "dt_rank": 82, "dtype": [1, 7, 10, 13, 14, 15, 16, 19, 70, 73, 75, 76, 82, 83, 84, 85, 86, 87, 92, 93, 98], "dual": 65, "due": [0, 12, 19, 23, 26, 28, 65, 73, 75, 79, 81, 87, 93, 97], "dummi": [70, 75, 93], "dump": [0, 3, 65, 70], "dump_debug_buff": 87, "duplic": [28, 93], "duplicate_data": 82, "durat": [0, 75], "duration_m": 70, "durationm": 0, "dure": [0, 1, 5, 6, 7, 11, 12, 13, 16, 24, 26, 27, 28, 29, 65, 70, 72, 73, 80, 81, 87, 89, 92, 97, 98], "dynam": [0, 26, 27, 29, 70, 73, 82, 84, 87, 89, 93, 99], "dynamic_batch_config": 70, "dynamic_batch_moving_average_window": 70, "dynamic_quant_bf16tonvfp4": 26, "dynamic_tree_max_topk": [43, 44, 70], "dynamicbatchconfig": [0, 70], "dynamicbatchmovingaveragewindow": 0, "dynamicbatchsizeconfig": 0, "dynamicdecodelay": 1, "dynamicqu": 26, "dynamictreemaxtopk": 0, "dynamictreemaxtopkhost": 1, "dynlibload": 0, "e": [0, 2, 3, 5, 8, 9, 10, 11, 17, 27, 28, 30, 31, 52, 55, 56, 57, 65, 70, 72, 82, 85, 87, 90, 92, 93, 95], "e2": [28, 64], "e4m3": [11, 22], "e5m2": 22, "each": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 15, 16, 20, 26, 27, 28, 29, 30, 36, 52, 55, 56, 57, 70, 73, 74, 75, 76, 79, 80, 81, 82, 83, 85, 87, 89, 90, 92, 93, 96, 97, 98, 99], "eager": [28, 71, 93], "eagl": [0, 1, 29, 40, 41, 43, 64, 70, 84, 87, 93], "eagle2": [40, 41], "eagle3_one_model": 70, "eagle_choic": [43, 44, 70, 87], "eagle_dynamic_tree_max_top_k": 87, "eagle_posterior_threshold": 87, "eagle_temperatur": 84, "eagle_use_dynamic_tre": 87, "eaglechoic": [0, 1], "eagleconfig": [0, 1, 84], "eagledecodingconfig": [43, 44, 70], "eagleforcausallm": 84, "eagleinput": 1, "eaglelastinput": 1, "eaglenetctxcontextlengthshost": 1, "eaglenetctxpastkeyvaluelengthshost": 1, "eaglenetctxrequesttypeshost": 1, "eaglenetgencontextlengthshost": 1, "eaglenetgenpastkeyvaluelengthshost": 1, "eaglenetgenrequesttypeshost": 1, "ealge2": 27, "earli": [87, 92, 93], "earlier": [0, 15, 77, 92], "early_stop": [6, 70, 87, 93], "early_stop_criteria": 87, "earlystop": [0, 1, 6], "eas": [18, 71, 74], "easi": [25, 75], "easier": [16, 19, 20, 27, 73], "easili": [17, 18, 20, 26, 71, 82], "east": [14, 16, 92], "eastern": 88, "ebnf": [0, 3, 70], "echo": [30, 31, 32, 56, 57], "eddi": 93, "edg": 22, "edit": [12, 65], "ef648e7489c040679d87ed12db5d3214": 88, "effect": [0, 2, 6, 11, 12, 26, 27, 28, 29, 67, 70, 77, 79, 80], "effici": [4, 5, 6, 9, 12, 16, 18, 26, 27, 28, 29, 34, 42, 46, 47, 49, 50, 61, 89, 91, 94, 96, 97, 98], "effort": [12, 15, 27, 28, 59, 77, 93], "eg": 74, "eight": [20, 21], "einop": 82, "einstein": 82, "einsum": 82, "einsum_eq": 82, "either": [0, 1, 2, 3, 18, 26, 28, 46, 49, 59, 70, 82, 89, 92, 93], "element": [0, 1, 5, 6, 10, 11, 70, 82, 83, 90], "element_typ": 1, "elementwis": [7, 82], "elementwise_affin": 83, "elementwise_binari": 82, "elementwise_sub": 7, "elementwise_sum": 7, "elementwiseoper": [7, 82], "eleutherai": 73, "elif": 99, "elimin": [2, 12, 26, 28, 29, 71, 73, 77, 79, 93], "ellipsi": 82, "els": [0, 16, 17, 19, 36, 52, 54, 59, 82, 92, 99], "elsinor": 51, "emb": [16, 61, 83], "embark": 71, "embed": [0, 9, 14, 27, 29, 70, 73, 82, 87, 93, 95, 97], "embed_dim": 83, "embed_posit": 83, "embed_positions_for_gpt_attent": 83, "embed_positions_for_gpt_attention_loc": 83, "embed_positions_loc": 83, "embed_token": [17, 95], "embedding_bia": 70, "embedding_dim": 83, "embedding_multipli": 84, "embedding_parallel_mod": 70, "embedding_scal": 84, "embedding_sharding_dim": [15, 84], "embeddingbia": [0, 1], "embeddingt": [0, 1], "emerg": [25, 26], "emit": 70, "emphasi": 15, "emploi": [12, 96, 99], "empow": 26, "empti": [0, 1, 12, 36, 82, 93, 99], "emptybuff": 1, "emptygenslot": 0, "emptytensor": 1, "emul": [82, 93], "en": 93, "enabl": [0, 2, 3, 5, 6, 7, 10, 11, 12, 13, 16, 17, 22, 23, 24, 25, 26, 27, 28, 29, 30, 32, 36, 42, 48, 50, 64, 65, 66, 67, 70, 73, 75, 79, 81, 82, 83, 84, 85, 87, 88, 90, 92, 93, 95, 97, 98], "enable_allreduc": 82, "enable_attention_dp": [20, 30, 56, 70], "enable_batch_size_tun": 70, "enable_block_reus": [30, 43, 44, 51, 54, 70], "enable_build_cach": [70, 93], "enable_chunked_context": [87, 93], "enable_chunked_prefil": [70, 93], "enable_context_fmha_fp32_acc": [70, 87], "enable_debug_output": [29, 70, 92], "enable_forward_chunk": 84, "enable_fp8": [11, 59], "enable_if_t": 1, "enable_iter_perf_stat": [30, 70], "enable_iter_req_stat": 70, "enable_kv_cache_reus": 9, "enable_layerwise_nvtx_mark": 70, "enable_lora": [58, 70], "enable_max_num_tokens_tun": [70, 93], "enable_min_lat": 70, "enable_multi_devic": 93, "enable_nvfp4": 59, "enable_overlap_schedul": 30, "enable_partial_reus": 70, "enable_prompt_adapt": [70, 93], "enable_qkv": 83, "enable_tqdm": 70, "enable_trt_overlap": 93, "enable_trtllm_sampl": 70, "enable_ucx": 93, "enable_xqa": 93, "enableattentiondp": [0, 1], "enablebatchsizetun": 0, "enableblockreus": [0, 9], "enablechunkedcontext": 0, "enablecontextfmhafp32acc": 0, "enabled_with_fp32_acc": 5, "enablelookaheaddecod": 1, "enablemaxnumtokenstun": 0, "enablepartialreus": 0, "enableseamlesslookaheaddecod": [0, 1], "enabletrtoverlap": 0, "enc": [29, 87, 93], "enc_dec": 6, "encapsul": [5, 6, 16, 82], "encdecmodelrunn": 87, "encod": [0, 5, 6, 22, 26, 29, 70, 82, 87, 90, 91, 93], "encode_base64_content_from_url": 61, "encoded_vocab": [0, 3], "encodedvocab": [0, 3], "encoder_hidden_st": [83, 84], "encoder_input_featur": 87, "encoder_input_id": 87, "encoder_input_len_rang": 93, "encoder_input_length": [82, 83, 87], "encoder_language_adapter_rout": 87, "encoder_max_input_length": [83, 87], "encoder_output": [83, 84, 87], "encoder_output_length": 87, "encoder_run": 87, "encoderenginebuff": 0, "encoderhiddens": 1, "encoderinputfeatur": 0, "encoderinputtokenid": 0, "encoderjsonconfigstr": 0, "encoderlen": 0, "encodermodel": [0, 84], "encodermodelpath": 0, "encoderoutput": 0, "encoderoutputlength": 0, "encount": [17, 20, 67, 92], "encourag": [0, 6, 19, 70], "end": [0, 1, 5, 6, 16, 27, 29, 43, 44, 49, 53, 54, 59, 70, 71, 73, 77, 80, 81, 82, 88, 93, 98], "end_dim": 82, "end_id": [70, 87, 93], "end_token": [0, 70], "endeavor": 26, "endid": [0, 1], "endpoint": [38, 39, 70, 88, 93], "endswith": 17, "enforc": [75, 82], "engin": [0, 1, 2, 3, 5, 6, 7, 10, 12, 13, 18, 19, 24, 26, 27, 28, 29, 30, 36, 46, 49, 58, 64, 67, 70, 74, 76, 77, 79, 80, 81, 82, 84, 87, 89, 92, 93], "engine_buff": 87, "engine_dir": [13, 14, 15, 16, 19, 70, 73, 75, 87, 88, 92], "engine_inspector": 87, "engine_llama_3": 16, "engine_nam": 87, "engine_output": 29, "engineaddr": 1, "enginebuff": [0, 1], "enginefilenam": 1, "engineinput": 1, "engineoutput": 1, "enginepath": 1, "engines": 1, "enhanc": [4, 6, 12, 20, 26, 27, 28, 71, 81, 89, 94, 97], "enjoi": [32, 42, 46, 47, 49, 50, 52], "enough": [5, 9, 20, 27, 79, 89, 96, 99], "enqueu": [0, 3, 16, 87, 89, 93], "enqueuecontext": 0, "enqueuegener": 0, "enqueuerequest": [0, 2, 3], "ensur": [2, 3, 4, 7, 19, 27, 65, 70, 73, 79, 85, 95, 98], "enter": [7, 31, 74, 79, 98], "enterpris": 45, "entir": [0, 3, 10, 16, 21, 26, 70, 71, 73, 74, 82, 89, 98], "entri": [0, 10, 40, 50, 66, 67, 73, 82, 88, 93], "entrypoint": [31, 69, 75], "enum": [0, 1, 2], "enumer": [0, 1, 48, 52, 94], "env": [30, 33, 34, 35, 37, 38, 39, 73], "envelop": 53, "environ": [6, 12, 20, 26, 34, 55, 56, 57, 61, 64, 65, 67, 72, 73, 75, 77, 79, 80, 92, 93, 94, 97], "environment": 17, "eo": [6, 70], "eof": [20, 27, 30, 56], "eos_token_id": [3, 87], "ep": [4, 20, 26, 27, 30, 73, 82, 83], "ep2": 26, "ep2tp4": 26, "ep4tp2": 26, "ep8": 28, "ep8tp8": 26, "ep_siz": [30, 37], "epsilon": [0, 82], "eq": 82, "equal": [0, 1, 3, 4, 28, 29, 36, 76, 82, 83, 89], "equal_progress": [70, 81], "equat": [24, 82], "equip": [2, 18], "equival": [26, 28, 77, 82, 95], "equvili": 29, "erenup": 93, "err": [55, 56, 57], "error": [0, 2, 3, 10, 19, 28, 29, 30, 59, 64, 65, 67, 70, 75, 79, 89, 93], "errorcod": 69, "errormsg": 0, "especi": [7, 27, 29, 42, 46, 47, 49, 50, 52, 76, 79, 98], "essenti": [12, 73], "establish": 28, "estim": [59, 73, 93, 99], "et": 21, "etc": [0, 1, 12, 67, 70, 72, 77, 80, 87, 89, 92, 95], "ethnzhng": 93, "eval": 45, "evalu": [11, 22, 23, 28, 64, 93], "even": [2, 5, 6, 16, 19, 25, 26, 29, 51, 75, 79, 82, 89], "evenli": [4, 26], "event": [0, 1, 40, 41, 64, 70], "event_buffer_max_s": [51, 70], "event_id": 51, "eventbuffermaxs": 0, "eventid": 0, "eventptr": 1, "ever": [0, 80], "everi": [0, 3, 17, 26, 28, 73, 75, 76, 82, 87], "everyon": 27, "everyth": 16, "evict": [0, 1, 8, 9, 10, 27, 71, 73, 75, 79], "evolv": [5, 19, 26, 71, 90, 98], "ex": [56, 57], "exact": [5, 89], "exam": 26, "examin": 12, "exampl": [0, 5, 6, 7, 9, 12, 13, 14, 18, 19, 21, 23, 25, 27, 30, 36, 45, 52, 55, 59, 64, 65, 69, 70, 74, 75, 76, 77, 78, 79, 80, 81, 82, 87, 88, 89, 90, 91, 92, 93, 94, 95, 97, 99], "example_logits_processor": 52, "exaon": [17, 91, 93], "exc": 48, "exce": [0, 2, 70, 81, 82], "exceed": [0, 89], "except": [0, 3, 5, 6, 19, 26, 27, 29, 54, 76, 82, 92, 93], "excess": 5, "exchang": 70, "excit": [42, 46, 47, 48, 49, 50], "exclud": [70, 77, 82, 93], "exclude_input_from_output": 70, "exclude_modul": [15, 70, 93], "excludeinputfromoutput": 0, "exclus": [1, 6, 90, 93], "exec": 72, "execut": [0, 2, 3, 6, 10, 12, 16, 18, 19, 26, 28, 64, 70, 71, 72, 73, 79, 81, 82, 87, 88, 89, 96, 99], "executor": [1, 2, 9, 12, 13, 18, 36, 52, 58, 64, 70, 71, 73, 81, 87, 89, 93, 96], "executor_config": 98, "executorconfig": [0, 3, 13], "executorexampledisaggreg": 2, "executorexamplefastlogit": 93, "exhaust": [0, 18], "exist": [1, 6, 9, 10, 12, 17, 19, 26, 28, 29, 51, 67, 70, 73, 87, 93, 97], "exit": [74, 87], "exp": 82, "expand": [0, 23, 25, 27, 82, 87, 93], "expand_dim": 82, "expand_dims_lik": 82, "expand_mask": 82, "expand_shap": 82, "expans": 82, "expect": [0, 5, 6, 11, 14, 16, 17, 19, 23, 27, 29, 36, 55, 56, 57, 64, 70, 73, 75, 78, 82, 92, 93], "expens": [3, 12, 71, 76, 77, 81], "experi": [12, 24, 25, 26, 28, 69, 71, 72, 73, 92], "experiment": [5, 6, 12, 17, 27, 30, 55, 56, 57, 64, 73, 90, 93, 94], "expert": [10, 20, 30, 50, 64, 70, 80, 93], "expertis": [26, 28], "expir": 0, "explain": [6, 16, 18, 28, 79, 82, 89, 90, 96, 97], "explan": [20, 28, 80, 87, 89], "explicit": [0, 1, 12, 82, 93], "explicit_draft_token": [12, 29, 84], "explicitdrafttoken": [0, 1], "explicitdrafttokensinput": 1, "explicitdrafttokenslastinput": 1, "explicitdrafttokensmodul": 1, "expliciteosstop": 0, "explicitli": [1, 2, 7, 12, 16, 17, 28, 29, 30, 36, 70, 93], "explor": [12, 26, 28, 71], "expon": 22, "exponenti": 12, "export": [2, 11, 15, 19, 20, 26, 27, 29, 30, 38, 39, 55, 56, 57, 73, 86, 87, 92, 93], "export_fmt": 94, "expos": [0, 6, 16, 32, 65, 77, 93], "express": [0, 3, 70, 82], "extend": [0, 3, 9, 16, 26, 27, 28, 70, 80, 82, 93], "extended_runtime_perf_knob_config": [70, 93], "extendedruntimeperfknobconfig": [0, 70], "extens": [15, 18, 67, 71, 73, 93], "extern": [0, 7, 8, 17, 87, 89], "external_checkpoint_dir": 17, "external_kei": 17, "external_weight": 17, "externaldrafttoken": 0, "externaldrafttokensconfig": [0, 1], "externaldrafttokensinput": 1, "externalstream": 52, "extra": [0, 2, 5, 9, 12, 15, 20, 26, 27, 29, 30, 37, 67, 70, 74, 76, 77, 87, 93], "extra_arg": 56, "extra_data": 70, "extra_id": 9, "extra_llm_api_opt": [20, 27, 30, 37, 56, 73, 74], "extra_resource_manag": 70, "extra_token": 83, "extract": [0, 3, 65, 72, 78, 82, 87], "extrapol": 82, "extrem": [16, 26, 77, 79, 80], "f": [0, 5, 6, 31, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 66, 67, 70, 72, 75, 81, 82, 88, 92, 94], "face": [3, 10, 13, 18, 19, 36, 70, 73, 84, 88, 93], "facilit": [7, 12, 88], "fact": [71, 73, 80], "factor": [25, 28, 76, 77, 82, 83, 89, 90], "factori": [19, 70, 87, 93], "factual": 6, "fail": [70, 87, 89, 92, 99], "failur": [17, 93], "fairli": 16, "fairseq": [91, 93], "fake": [9, 93], "fakebuff": 1, "falcon": [15, 25, 69, 73, 90, 91, 93], "falconconfig": 84, "falconforcausallm": 84, "falconmodel": 84, "fall": [11, 67, 74, 93], "fallback": 17, "fals": [0, 1, 2, 3, 5, 6, 7, 9, 15, 26, 28, 29, 30, 45, 51, 54, 56, 70, 82, 83, 84, 85, 86, 87, 93], "false_output_valu": 82, "false_valu": 82, "famili": [5, 17, 91, 93], "familiar": [6, 16, 69, 75, 76, 78, 88], "famou": [6, 59], "faq": 64, "far": [0, 3, 27], "fast": [0, 5, 8, 12, 70, 73, 76, 93], "fast_build": [29, 70, 93], "fastapi": 93, "fastapi_serv": 93, "faster": [5, 19, 22, 23, 27, 28, 29, 74, 75, 82], "fasterdecod": 54, "fastlogit": 0, "fault": 93, "favor": 93, "favorit": 58, "fc": [15, 16, 17, 92], "fc_gate": 83, "fc_gate_dora": 83, "fc_gate_lora": 83, "fc_gate_plugin": 83, "featur": [0, 2, 3, 5, 7, 8, 10, 11, 12, 15, 16, 17, 19, 25, 26, 27, 28, 29, 55, 56, 57, 64, 65, 73, 77, 79, 80, 81, 82, 85, 87, 91, 94, 95, 97], "feature_dim": 87, "februari": 28, "fed": [74, 84], "feed": 82, "feedback": 93, "feedforward": 4, "feel": 58, "fetch": [0, 27, 30, 96], "few": [9, 16, 19, 25, 28, 79], "fewer": [5, 12, 21, 97], "ffn": [4, 26], "ffn_hidden_s": 83, "fhma": 93, "field": [0, 6, 15, 19, 30, 32, 36, 70, 71, 73, 77, 84, 85, 90, 93, 97], "field_nam": 70, "figur": [26, 27], "file": [0, 3, 4, 5, 7, 9, 15, 16, 17, 19, 20, 27, 29, 30, 38, 39, 67, 70, 72, 73, 74, 87, 88, 93, 95], "filepath": 1, "filesystem": [0, 1], "fill": [1, 17, 32, 42, 46, 47, 49, 50, 82, 97], "fill_attention_const_params_for_long_rop": 83, "fill_attention_const_params_for_rop": 83, "fill_attention_param": 83, "fill_none_tensor_list": 83, "fill_valu": [52, 82], "fillemptyfieldsfromruntimedefault": 0, "filloper": 82, "filltaskstensor": 1, "filter_medusa_logit": 87, "final": [0, 1, 10, 26, 27, 29, 30, 31, 36, 82, 99], "final_logit_softcap": 84, "final_output_id": 87, "finalize_decod": 87, "find": [20, 28, 77, 82, 92, 93], "find_best_medusa_path": 87, "fine": [12, 20, 28, 73, 80, 83], "finer": 7, "finetun": 26, "finish": [0, 1, 3, 6, 8, 19, 27, 36, 53, 69, 70, 71, 73, 87, 96, 98], "finish_reason": [53, 70, 88, 93], "finishedst": 1, "finishedsum": 1, "finishreason": [0, 1, 93], "first": [0, 1, 2, 3, 5, 6, 7, 9, 10, 12, 18, 23, 25, 27, 28, 29, 30, 31, 67, 69, 70, 73, 74, 75, 77, 79, 80, 81, 82, 89, 92, 93, 95, 97, 98, 99], "first_come_first_serv": [70, 81], "first_gen_token": 70, "first_lay": 87, "firstgentoken": 0, "firstit": 0, "firstli": [28, 31, 79, 89], "firstscheduledtim": 0, "firsttokentim": 0, "fit": [1, 5, 21, 22, 70, 76, 77, 99], "fitting_request": 99, "fix": [8, 10, 12, 28, 73, 89], "fjosw": 93, "flag": [0, 1, 3, 5, 10, 19, 24, 30, 36, 64, 73, 77, 78, 79, 81, 82, 89, 93], "flags_siz": 1, "flan": [90, 91], "flash": [5, 16], "flashattent": [5, 16, 88], "flashinf": 97, "flashinferattent": 97, "flashmla": [27, 93], "flatten": [1, 10, 24, 82, 83], "flattenedinouts": 1, "flattenn": 1, "flayer": 7, "flayerinfomemo": 7, "flexibl": [12, 19, 26, 36, 65], "flight": [1, 18, 64, 73, 79, 81, 88, 89, 93], "flip": 82, "flip_sin_to_co": 83, "float": [0, 1, 6, 13, 15, 16, 22, 52, 70, 81, 82, 83, 84, 87, 90], "float16": [7, 10, 13, 14, 15, 19, 29, 76, 82, 84, 85, 88, 92], "float2": 82, "float32": [0, 15, 29, 82, 83, 84, 85], "floattensor": 95, "floattyp": [0, 1], "floor_div": 82, "floordiv": 82, "flop": 28, "flow": [7, 19, 26, 28, 75, 76, 77, 79, 80, 93, 96, 99], "fly": [5, 82, 90], "fmha": [0, 29, 70, 82, 87, 89, 93], "fmt_dim": 1, "focu": [7, 25, 26, 72], "focus": [12, 73, 77, 78, 93], "fold": 89, "folder": [0, 3, 6, 19, 75, 90, 91, 93], "folder_trt_llm": 16, "follow": [3, 6, 7, 10, 12, 14, 15, 16, 17, 19, 20, 25, 26, 27, 28, 29, 30, 32, 36, 47, 48, 51, 55, 56, 57, 65, 66, 67, 69, 73, 74, 75, 76, 77, 78, 79, 80, 82, 88, 90, 91, 93, 94, 95, 97, 98], "footprint": [5, 21, 28, 89], "for_each_rank": 84, "forc": [0, 5, 11, 26, 73], "force_drop_id": 83, "force_low_precision_all_reduce_strategi": 11, "force_multi_block_mod": 73, "force_nccl_all_reduce_strategi": 93, "force_num_profil": 70, "force_words_id": 6, "forecast": 12, "foretel": 51, "fork": 72, "form": [0, 3, 5, 12, 70, 82, 88], "format": [0, 3, 11, 15, 17, 19, 22, 25, 27, 28, 39, 64, 65, 69, 70, 71, 75, 77, 87, 88, 89, 92, 93, 97], "former": [16, 25, 51], "formula": [28, 82], "forum": 93, "forward": [0, 1, 7, 12, 14, 16, 27, 52, 81, 82, 83, 84, 92, 93, 95, 96, 97, 98, 99], "forward_loop": 73, "forward_with_cfg": 84, "forward_without_cfg": 84, "forwardasync": 1, "forwarddispatch": 1, "forwardsync": 1, "found": [3, 4, 5, 6, 7, 12, 16, 18, 22, 65, 67, 73, 75, 77, 80, 90, 99], "foundat": 27, "four": [3, 7, 12, 15, 26, 27, 83], "fourth": 3, "fp": [90, 93], "fp16": [5, 10, 11, 13, 15, 17, 21, 22, 25, 29, 64, 73, 77, 80, 82, 88, 91, 92, 93], "fp32": [0, 5, 26, 28, 29, 64, 70, 82, 87, 88, 91, 92, 93], "fp4": [20, 27, 28, 29, 93], "fp8": [11, 19, 20, 21, 23, 24, 25, 26, 27, 28, 29, 46, 54, 59, 64, 70, 73, 78, 80, 82, 85, 89, 91, 93, 94, 97], "fp8_block_scal": 70, "fp8_blockscale_gemm": 93, "fp8_inputs_overrid": 82, "fp8_kv_cach": [5, 90], "fp8_per_channel_per_token": 70, "fp8_qdq": 90, "fp8_rowwise_gemm_plugin": 29, "fp_valu": 5, "fpa_intb": 93, "fraction": [0, 30, 70, 82, 83, 87], "framework": [12, 14, 15, 18, 19, 71, 82, 93], "franc": [14, 16, 40, 42, 43, 44, 46, 47, 48, 49, 50, 54, 59, 66, 67, 75, 81, 88, 92, 94], "free": [0, 1, 8, 10, 16, 17, 28, 30, 71, 79, 83, 84, 87, 89, 98], "free_gpu_memory_fract": [30, 36, 49, 53, 70, 81, 93], "free_resourc": [96, 98], "freed": 73, "freedom": 19, "freegpumemoryfract": [0, 89, 93], "freenumblock": 0, "freez": 28, "french": 88, "freq": 82, "frequenc": [73, 83], "frequency_penalti": [70, 87, 93], "frequencypenalti": [0, 1, 6], "frequent": [9, 92], "friend": [0, 1, 73], "friendli": 82, "from": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 23, 25, 26, 27, 28, 29, 30, 31, 32, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 64, 66, 67, 69, 70, 71, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 84, 86, 87, 88, 89, 92, 93, 94, 95, 96, 97, 98, 99], "from_argu": 84, "from_checkpoint": [19, 84], "from_config": 84, "from_dict": [70, 84], "from_dir": 87, "from_engin": 87, "from_hugging_fac": [14, 17, 19, 84], "from_jax": 19, "from_json_fil": [70, 84], "from_kera": 19, "from_meta_ckpt": [19, 84], "from_nemo": [19, 84], "from_pretrain": 84, "from_prun": 84, "from_serialized_engin": 87, "from_str": 82, "fromfil": 16, "fruit": 28, "full": [0, 4, 5, 6, 9, 10, 12, 22, 23, 27, 28, 30, 70, 71, 72, 73, 76, 82, 87, 88, 89, 92], "full_lik": 52, "fulli": [28, 40, 93], "funcnam": 0, "function": [0, 1, 3, 5, 13, 14, 16, 18, 19, 26, 27, 69, 70, 71, 72, 80, 85, 87, 89, 90, 91, 92, 93, 98, 99], "functiont": 0, "further": [3, 4, 5, 12, 16, 21, 25, 27, 28, 29, 73, 77, 80, 97], "furthermor": [12, 26, 77], "fuse": [5, 12, 16, 26, 28, 29, 80, 82, 88, 93, 95, 97], "fuse_a": [26, 28], "fuse_fp4_qu": 29, "fuse_qkv_project": 84, "fuseattentionwithbiaspass": 7, "fused_gate_up_dora": 83, "fused_gate_up_lora": 83, "fused_mo": 70, "fusedgatedmlp": [82, 83], "fusevalu": 1, "fusion": [7, 28, 29, 64, 71, 79, 89, 90, 93, 97], "fusion_op": 82, "futur": [2, 5, 6, 8, 12, 17, 19, 25, 29, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 54, 59, 65, 66, 67, 69, 70, 71, 73, 75, 81, 82, 88, 89, 90, 93, 94], "fuyu": [91, 93], "g": [3, 8, 11, 17, 27, 28, 30, 52, 55, 56, 57, 70, 79, 87, 95], "g1": 79, "g2": 79, "gain": [76, 79], "gamma": 82, "gate": [10, 17, 29, 75, 82, 93], "gate_a": 82, "gate_a_bia": 82, "gate_bia": 82, "gate_proj": 17, "gate_x": 82, "gate_x_bia": 82, "gatedmlp": [82, 83], "gather": [0, 1, 29, 47, 48, 70, 82, 87], "gather_all_token_logit": [29, 93], "gather_context_logit": [29, 70, 84, 87], "gather_dim": [16, 82], "gather_generation_logit": [29, 70, 84, 87], "gather_last_token_logit": 82, "gather_nd": 82, "gather_output": 83, "gathercontext": [0, 93], "gatheredid": 1, "gatherel": 82, "gathergenerationlogit": 0, "gathermod": 82, "gathertre": 1, "gatherv2": 82, "gb": [2, 23, 28, 65, 70, 73], "gb200": [28, 93], "gcc": 65, "gd": 0, "gdrdma": 2, "geforc": 93, "gegelu": 82, "gegelu_limit": 83, "geglu": 82, "gelu": [82, 84], "gelu_pytorch_tanh": 93, "gelu_tanh": 83, "gemm": [7, 28, 29, 79, 82, 88, 89, 93], "gemm_allreduc": 82, "gemm_allreduce_plugin": [29, 87], "gemm_fc1": 26, "gemm_plugin": [10, 13, 15, 16, 29, 73, 77, 80, 83, 88], "gemm_swiglu": 82, "gemm_swiglu_plugin": [29, 77, 85], "gemma": [19, 69, 90, 91, 93], "gemma2": 91, "gemma2_added_field": 84, "gemma2_config": 84, "gemma3": 93, "gemma3_added_field": 84, "gemma3_config": 84, "gemma_added_field": 84, "gemma_config_kwarg": 84, "gemmaconfig": 84, "gemmaforcausallm": 84, "gen": [70, 93], "genai": [25, 30, 63], "genattent": 26, "genenginepath": 0, "gener": [0, 1, 3, 6, 9, 12, 15, 16, 17, 19, 20, 21, 22, 24, 26, 27, 28, 29, 40, 41, 42, 51, 64, 66, 67, 69, 70, 71, 72, 73, 74, 75, 76, 78, 79, 80, 81, 82, 84, 87, 88, 89, 91, 92, 93, 94, 95, 96, 97, 98, 99], "generate_alibi_bias": 82, "generate_alibi_slop": 82, "generate_async": [36, 47, 48, 70, 93], "generate_logn_sc": 82, "generate_tllm_weight": 17, "generated_text": [40, 43, 44, 54, 58, 66, 67, 75, 81, 88, 94], "generatedtokensperenginestep": 1, "generation_complet": 99, "generation_in_progress": 99, "generation_logit": [53, 70, 87], "generation_onli": 70, "generation_phas": 5, "generation_request": 99, "generation_to_complet": 99, "generationexecutor": [2, 93], "generationlength": 1, "generationlengthsdevic": 1, "generationlengthshost": 1, "generationlengthshostcopi": 1, "generationlogit": 0, "generationmixin": 84, "generationrequestid": 2, "generationresult": 70, "generationsequ": 87, "generationsess": [5, 87, 89], "generationstep": 1, "genericprompttuningparam": 1, "genert": 2, "genexecutorconfig": 0, "genidx": 0, "genrequest": 1, "genrespons": 2, "get": [0, 1, 2, 3, 5, 7, 10, 13, 17, 24, 27, 28, 30, 31, 32, 36, 40, 41, 65, 66, 67, 70, 71, 72, 75, 77, 82, 84, 87, 88, 92, 93, 94, 99], "get_1d_sincos_pos_embed_from_grid": 83, "get_2d_sincos_pos_emb": 83, "get_2d_sincos_pos_embed_from_grid": 83, "get_audio_featur": 87, "get_batch_cache_indic": 98, "get_batch_idx": 87, "get_block_offset": 87, "get_buff": 98, "get_comm": 70, "get_config_group": 84, "get_context_phase_param": 70, "get_device_cap": 59, "get_first_past_key_valu": 83, "get_hf_config": 84, "get_input": 7, "get_kv_cache_ev": [51, 70], "get_kv_cache_events_async": 70, "get_max_resource_count": [98, 99], "get_needed_resource_to_complet": [98, 99], "get_next_medusa_token": 87, "get_num_free_block": 98, "get_num_heads_kv": 87, "get_output": [7, 16], "get_par": [7, 82], "get_pytorch_backend_config": 70, "get_request_typ": 70, "get_rope_index": 87, "get_seq_idx": 87, "get_shap": 17, "get_slic": 17, "get_stat": [70, 93], "get_stats_async": 70, "get_timestep_embed": 83, "get_us": [7, 82], "get_visual_featur": 87, "get_vocab": [0, 3], "get_weight": 83, "getacceptancethreshold": 0, "getacceptedlengthscumsum": 1, "getacceptedpackedpath": 1, "getadditionalmodeloutput": 0, "getadditionaloutputnam": 0, "getaddr": 0, "getaddress": 1, "getagentst": 0, "getallnewtoken": 1, "getallottedtimem": 0, "getattentionconfig": 0, "getbackend": 0, "getbackendagentdesc": 0, "getbadword": 0, "getbatchingtyp": 0, "getbatchsizet": 0, "getbeamsearchbuff": 1, "getbeamsearchdiversityr": 0, "getbeamwidth": 0, "getbeamwidtharrai": 0, "getbuffermanag": 1, "getcachest": 0, "getcachetransceiverconfig": 0, "getcapac": 1, "getcapacityschedulerpolici": 0, "getclientid": 0, "getcommptr": 1, "getcommst": 0, "getcommunicationmod": 0, "getcommunicationtyp": 0, "getconfig": 0, "getconnect": 0, "getconnectioninfo": 0, "getcontextchunkingpolici": 0, "getcontextexecutor": 0, "getcontextfmha": 1, "getcontextparallel": 1, "getcontextparallelgroup": 1, "getcontextparallelrank": 1, "getcontextphaseparam": 0, "getcopyonpartialreus": 0, "getcpu": 1, "getcpudiff": 1, "getcrossattentionmask": 0, "getcrosskvcachefract": 0, "getcudagraphcaches": 0, "getcudagraphmod": 0, "getcumlogprob": 1, "getdata": 0, "getdatatyp": [0, 1], "getdatatypenam": 1, "getdebugconfig": 0, "getdebuginputtensor": 0, "getdebugoutputtensor": 0, "getdebugtensornam": 0, "getdebugtensorsmaxiter": 0, "getdecodedurationm": 0, "getdecoderetentionprior": 0, "getdecoderst": 1, "getdecoderstream": 1, "getdecodingconfig": 0, "getdecodingmod": 0, "getdefaultbatchslot": 1, "getdefaulteaglechoic": 1, "getdesc": 0, "getdevic": 1, "getdevicecacheperc": 0, "getdeviceid": 0, "getdeviceof": 1, "getdimens": 1, "getdirectori": 0, "getdrafttoken": 0, "getdstdesc": 0, "getdynamicbatchconfig": 0, "getdynamicbatchmovingaveragewindow": 0, "getdynamictreemaxtopk": 0, "geteaglechoic": 0, "geteagleconfig": 0, "getearlystop": 0, "getembeddingbia": 0, "getembeddingt": 0, "getenablebatchsizetun": 0, "getenableblockreus": 0, "getenablechunkedcontext": 0, "getenablecontextfmhafp32acc": 0, "getenablemaxnumtokenstun": 0, "getenablepartialreus": 0, "getenabletrtoverlap": 0, "getencodedvocab": 0, "getencoderhiddens": 1, "getencoderinputfeatur": 0, "getencoderinputtokenid": 0, "getencoderoutputlength": 0, "getendid": 0, "geterrormsg": 0, "geteventbuffermaxs": 0, "getexecutionconfig": 1, "getextendedruntimeperfknobconfig": 0, "getexternaldrafttokensconfig": 0, "getfastlogit": 0, "getfinishedstep": 1, "getfinishedsum": 1, "getfinishreason": 1, "getfirstgentoken": 0, "getfirstlocallay": 1, "getfreegpumemoryfract": 0, "getfrequencypenalti": 0, "getfunctionpoint": 0, "getgatheredid": 1, "getgathergenerationlogit": 0, "getgemmallreducedtyp": 1, "getgenexecutor": 0, "getgpu": 1, "getgpudiff": 1, "getgpuspergroup": 1, "getgpuspernod": 1, "getgpuweightsperc": [0, 13], "getguid": 0, "getguideddecodingconfig": 0, "getguideddecodingparam": 0, "getguidetyp": 0, "gethandl": 0, "gethiddens": 1, "gethostcaches": 0, "gethostmemori": 1, "getid": 1, "getinittozero": 1, "getinputtokenextraid": 0, "getinputtokenid": 0, "getinst": [0, 1], "getipcunicastpoint": 1, "getisorchestr": 0, "getiterstatsmaxiter": 0, "getjointdecodinginput": 1, "getjointdecodingoutput": 1, "getkvcacheconfig": 0, "getkvcacheconfigref": 0, "getkvcacheeventmanag": 0, "getkvcacheretentionconfig": 0, "getkvcachetyp": 1, "getkvdatatyp": 1, "getlanguageadapteruid": 0, "getlastrank": 1, "getlatestdebugtensor": 0, "getlatestev": 0, "getlatestiterationstat": [0, 3], "getlatestrequeststat": 0, "getlayertyp": 1, "getlen": 0, "getlengthpenalti": 0, "getlevel": 1, "getlocalagentdesc": 0, "getlocalrank": 1, "getlogit": 0, "getlogitsdtyp": 1, "getlogitspostprocessor": 0, "getlogitspostprocessorconfig": 0, "getlogitspostprocessornam": 0, "getlogprob": 1, "getlookaheadconfig": 0, "getlookaheaddecodingconfig": 0, "getlookaheaddecodingmaxnumrequest": 0, "getloraconfig": 0, "getloramodul": 1, "getloraprefetchdir": 0, "getmanagedweightsmapopt": 1, "getmanageweightstyp": 1, "getmaxadapters": 0, "getmaxattentionwindowvec": 0, "getmaxbatchs": [0, 1], "getmaxbeamwidth": [0, 1], "getmaxdecodingdecodertoken": 1, "getmaxdecodingdrafttoken": 1, "getmaxdecodingenginetoken": 1, "getmaxdecodingtoken": 1, "getmaxdraftpathlen": 1, "getmaxencoderlen": 1, "getmaxinputlen": 1, "getmaxlorarank": 1, "getmaxnonleafnodesperlay": 1, "getmaxnumpath": 1, "getmaxnumtoken": [0, 1], "getmaxpagesperblock": 1, "getmaxpagesperblockdevic": 0, "getmaxpagesperblockhost": 0, "getmaxpathlen": 1, "getmaxpositionembed": 1, "getmaxpromptembeddingtables": 1, "getmaxqueues": 0, "getmaxseqidlemicrosecond": 0, "getmaxsequencelen": 1, "getmaxsequencelength": 1, "getmaxtoken": 0, "getmedusachoic": [0, 1], "getmemorytyp": [0, 1], "getmemorytypenam": 1, "getminp": 0, "getmintoken": 0, "getmlphiddens": 1, "getmodelconfig": [0, 1], "getmodelconfigmut": 1, "getmodelnam": 1, "getmodelvari": 1, "getmpist": 0, "getmropeconfig": 0, "getmropepositiondelta": 0, "getmroperotarycossin": 0, "getmultiblockmod": 0, "getmulticastpoint": 1, "getmultimodalembed": 0, "getnam": [0, 1], "getnbattentionlay": 1, "getnbhead": 1, "getnbkvhead": 1, "getnblay": 1, "getnbrnnlay": 1, "getnextdrafttoken": 1, "getnextdrafttokenslength": 1, "getngrams": 0, "getnoderank": 1, "getnoderankof": 1, "getnorepeatngrams": 0, "getnormalizelogprob": 0, "getnotifiedsyncmessag": 0, "getnumcopystream": [0, 1], "getnumdecodingenginetoken": 1, "getnumdevicemodulelay": 0, "getnumensurework": 0, "getnumhostmodulelay": 0, "getnumkvheadsperlay": 1, "getnumkvheadsperlayerlocalrang": 1, "getnumlanguag": 1, "getnumnod": 0, "getnumpackedmask": 1, "getnumpag": 1, "getnumputwork": 0, "getnumresponsesreadi": 0, "getnumreturnbeam": [0, 1], "getnumreturnsequ": 0, "getnumtransformerlay": 1, "getonboardblock": 0, "getop": 0, "getoptimaladapters": 0, "getoptprofilessplitpoint": 1, "getorchestratorconfig": 0, "getorchleadercomm": 0, "getoutputconfig": 0, "getpadid": 0, "getpagedcontextfmha": 1, "getpageptr": 1, "getpagewidth": 1, "getparallelconfig": 0, "getparentid": 1, "getparticipantid": 0, "getpath": 1, "getpathopt": 1, "getpeftcacheconfig": 0, "getperblockretentionprioritydur": 0, "getpin": 1, "getpinneddiff": 1, "getpinnedpool": 1, "getpinnedpooldiff": 1, "getpipelineparallel": 1, "getpipelineparallelgroup": 1, "getpipelineparallelrank": 1, "getpositionid": 0, "getposteriorthreshold": 0, "getppreducescatt": 1, "getprecis": 1, "getpresencepenalti": 0, "getprevdrafttokenslength": 1, "getprior": 0, "getprocessorbatch": 0, "getprocessormap": 0, "getprompttableoffload": 0, "getprompttuningconfig": 0, "getquantmod": 1, "getrank": 1, "getrecvpollperiodm": 0, "getremotenam": 0, "getrepetitionpenalti": 0, "getrepl": 0, "getreqid": 0, "getrequestid": 0, "getrequeststatsmaxiter": 0, "getrequesttyp": 0, "getresult": [0, 2, 3], "getreturnallgeneratedtoken": 0, "getrnnconfig": 1, "getrotaryembeddingdim": 1, "getruntimedefault": 1, "getruntimetyp": 0, "getsamplingconfig": [0, 1], "getschedulerconfig": 0, "getschedulerconfigref": 0, "getse": 0, "getsecondaryoffloadminprior": 0, "getselfidx": 0, "getsequencelength": 1, "getserializedst": 0, "getshap": [0, 1], "getsinktokenlength": 0, "getsiz": [0, 1], "getsizeinbit": 1, "getsizeinbyt": [0, 1], "getsizeperhead": 1, "getskipcrossattnblock": 0, "getslotsperpag": 1, "getsocketst": 0, "getspawnprocess": 0, "getspecdecconfig": 0, "getspeculativedecodingmod": 1, "getspeculativedecodingmodul": 1, "getspeculativedecodingmoduleptr": 1, "getsrcdesc": 0, "getstat": 0, "getstatu": 1, "getstoptokenid": 0, "getstopword": 0, "getstream": [0, 1], "getsumlocalkvhead": 1, "getsyncmessag": 0, "gettag": 0, "gettaskid": 0, "gettemperatur": 0, "gettensorparallel": 1, "gettensorparallelgroup": 1, "gettensorparallelrank": 1, "getter": 6, "gettoken": 0, "gettokenizerstr": 0, "gettokenrangeretentionconfig": 0, "gettokensperblock": 1, "gettopk": 0, "gettopp": 0, "gettoppdecai": 0, "gettoppmin": 0, "gettoppresetid": 0, "gettotalnumpag": 1, "gettransfermod": 0, "gettyp": [0, 1], "getunderlyingdecod": 1, "getunicastpoint": 1, "getusegpudirectstorag": 0, "getuvm": 1, "getuvmdiff": 1, "getverificationsets": 0, "getvers": 1, "getvocabs": 1, "getvocabsizepad": 1, "getweight": 0, "getwindows": 0, "getworkerexecutablepath": 0, "getworlds": 1, "gh200": 93, "ghost": 51, "ghz": 45, "gib": [9, 89], "gid": 0, "gigabyt": 23, "git": [10, 20, 65, 69, 88, 92, 94], "github": [19, 20, 26, 65, 69, 71, 88, 93, 94], "give": [3, 27, 28, 71, 77, 79, 84], "given": [0, 1, 3, 6, 10, 17, 19, 23, 69, 70, 72, 78, 79, 82, 83, 84, 86, 87, 89, 90, 93, 98], "givyboi": 58, "glm": [69, 82, 91, 93], "glm4": [69, 93], "global": [0, 5, 8, 16, 26, 28, 93], "global_max_input_length": 87, "global_max_output_length": 87, "globalrequestid": 0, "glossari": [21, 24], "gm": 92, "gnu": 65, "go": [5, 6, 51, 76, 93], "goal": 81, "goe": [27, 69, 73], "good": [3, 16, 20, 28, 73, 76, 79, 80], "got": [0, 42, 45, 46, 47, 48, 49, 50, 51, 52, 58, 59, 69, 73, 92], "gpqa": [26, 28], "gpt": [1, 5, 12, 16, 18, 22, 25, 29, 64, 69, 73, 82, 89, 90, 91, 92, 93], "gpt2": [84, 92], "gpt3": 23, "gpt_attent": [5, 7, 24, 82, 88, 93], "gpt_attention_plugin": [10, 16, 29, 73, 83, 87, 92, 93], "gpt_attention_plugin_remove_pad": 7, "gpt_variant": [84, 93], "gptattent": 7, "gptattentionpluginremovepaddingrewritepass": 7, "gptconfig": 84, "gptdecod": 6, "gptdecoderbatch": 93, "gptdecoderptr": 1, "gptforcausallm": 84, "gptj": 84, "gptjconfig": 84, "gptjforcausallm": 84, "gptjmodel": 84, "gptlmheadmodel": 92, "gptmanag": 93, "gptmanagerbenchmark": [9, 65, 93], "gptmodel": 84, "gptmodelconfig": 93, "gptneoxforcausallm": 84, "gptneoxmodel": 84, "gptq": [25, 64, 91, 93], "gptsession": 93, "gptsessionbenchmark": 93, "gpu": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 15, 18, 19, 22, 23, 24, 25, 27, 29, 30, 36, 55, 56, 57, 59, 64, 65, 67, 69, 70, 74, 75, 76, 77, 80, 82, 84, 87, 88, 91, 92, 93, 96, 97], "gpu_weights_perc": [13, 87], "gpudirect": 0, "gpumemusag": [0, 30], "gpus_per_nod": [29, 30, 70], "gpuspernod": [1, 6], "gpusync": 1, "gpuweightsperc": [0, 13], "gqa": [5, 8, 21, 24, 29, 82, 93, 97], "grace": [9, 64, 91], "gradient": 22, "gradual": 19, "grain": 7, "gram": 12, "grammar": [0, 3, 70], "granit": [91, 93], "graph": [0, 16, 20, 27, 28, 64, 70, 72, 73, 82, 87, 88, 89, 92, 93, 97, 98], "graph_rewrit": 7, "graphic": 53, "gratitud": 27, "gre": 30, "great": [21, 53], "greater": [0, 2, 5, 24, 25, 26, 29, 82], "greatli": [9, 19, 27, 77, 80], "greedi": [0, 6, 96], "greedy_sampl": [43, 44, 70], "greedysampl": 0, "greedysamplinghost": 1, "grid": [16, 77, 79, 82, 83], "grid_search_engin": 75, "grid_siz": 83, "grok": [91, 93], "ground": 72, "groundbreak": 71, "group": [0, 3, 4, 6, 8, 16, 21, 28, 64, 70, 82, 83, 90, 93, 97], "group_cl": 84, "group_norm": 82, "group_siz": [15, 70, 82], "groupedrmsnorm": 26, "groupgemm": 28, "groupnorm": [82, 83], "grow": [1, 12, 79], "gsm8k": 28, "gt": 82, "gtc": [20, 26], "guarante": [0, 6, 9, 19, 73, 74, 75, 77, 81], "guaranteed_no_evict": [0, 70, 73, 81], "guaranteednoevictschedul": 99, "guard": [51, 75], "guid": [0, 16, 20, 25, 40, 41, 64, 69, 70, 71, 72, 74, 75, 76, 77, 80, 82, 92, 93, 97], "guidanc": [12, 30, 80, 83, 84], "guided_decod": [45, 70], "guided_decoding_backend": [45, 70], "guideddecodingbackend": 0, "guideddecodingconfig": [0, 3], "guideddecodingparam": [0, 3, 45, 70], "guidelin": [2, 76], "guidetyp": [0, 3], "gw": 7, "h": [2, 3, 5, 12, 17, 27, 29, 30, 33, 34, 35, 75, 82, 84, 88, 93], "h0": 27, "h1": 82, "h100": [19, 25, 29, 71, 74, 75, 77, 78, 79, 93], "h20": 29, "h200": [22, 29, 74, 93], "h2d": 52, "ha": [0, 1, 3, 5, 9, 10, 11, 15, 16, 17, 19, 20, 21, 25, 26, 27, 28, 29, 32, 65, 70, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 87, 89, 90, 92, 93, 96, 98, 99], "had": [19, 28, 77, 79], "half": [0, 1, 16, 28, 75, 82], "halv": [22, 82], "hand": [9, 12, 18, 76], "handl": [0, 1, 2, 4, 8, 17, 19, 21, 26, 75, 77, 79, 80, 81, 82, 83, 95, 96], "handle_per_step": 87, "hang": [0, 69, 92, 93], "happen": [3, 6, 9, 16, 67, 89, 92], "happi": 87, "har": 28, "hard": [5, 70], "harder": 6, "hardwar": [8, 25, 28, 36, 64, 65, 93], "has_affin": 82, "has_bia": 82, "has_config_group": 84, "has_position_embed": 87, "has_scal": 82, "has_token_type_embed": 87, "has_zero_point": [15, 70], "hascontextawaitthread": 0, "hasdraftlogit": 1, "haserror": [0, 3], "hasgenawaitthread": 0, "hash": [0, 70], "hasresult": 0, "hasrnnconfig": 1, "hasspeculativedecodingmodul": 1, "hattizai": 93, "have": [0, 1, 3, 4, 5, 6, 9, 10, 12, 15, 16, 17, 19, 20, 21, 23, 25, 26, 27, 28, 29, 31, 51, 54, 55, 56, 57, 58, 67, 69, 70, 71, 72, 73, 75, 76, 77, 78, 79, 80, 81, 82, 87, 88, 89, 91, 92, 93, 95], "hbm3": 74, "hbm3e": 23, "he": 51, "head": [1, 6, 8, 12, 16, 21, 27, 28, 29, 54, 59, 64, 73, 82, 83, 93, 97], "head_dim": [97, 98], "head_siz": [5, 82, 84, 87, 93], "header": 2, "headsiz": 82, "headsperlay": 1, "health": [30, 58], "heat": 6, "heavi": 80, "heavier": 76, "height": [39, 83, 87], "hello": [40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 55, 58, 59, 66, 67, 75, 81, 88, 94], "help": [2, 3, 5, 7, 16, 26, 27, 29, 30, 33, 34, 45, 52, 54, 60, 61, 65, 72, 73, 74, 75, 78, 79, 80, 81, 82, 88, 93, 96], "helper": [1, 82], "henc": 95, "here": [2, 3, 7, 10, 13, 14, 15, 16, 17, 19, 20, 22, 23, 27, 28, 30, 32, 36, 40, 45, 65, 72, 75, 76, 77, 79, 80, 82, 87, 88, 89, 90, 92, 94, 97, 98, 99], "heterogen": 2, "heurist": [5, 28, 73, 82, 93], "hf": [6, 10, 13, 17, 29, 30, 46, 47, 48, 49, 50, 54, 55, 56, 57, 59, 73, 74, 75, 87, 91, 92, 94], "hf_config_or_dir": 84, "hf_lora_convert": 10, "hf_model": [73, 84], "hf_model_dir": [13, 14, 15, 19, 84], "hf_model_nam": 73, "hf_model_or_dir": 84, "hf_quant_config": 73, "hf_token": 73, "hfconfigordir": 84, "hgx": 23, "hi": 10, "hidden": [0, 3, 4, 5, 6, 10, 12, 26, 27, 70, 82, 83, 93], "hidden_act": [15, 83, 84], "hidden_dim": [0, 5, 82], "hidden_dim_per_head": [5, 82], "hidden_dtyp": 83, "hidden_s": [0, 7, 15, 17, 82, 83, 84, 87, 95, 97], "hidden_size_in": 10, "hidden_size_out": 10, "hidden_size_per_head": 82, "hidden_st": [14, 82, 83, 84, 87, 92, 95], "hidden_states_for_emb": 84, "hiddens": [0, 1, 6], "hide": [26, 28], "hierarch": 15, "hierarchi": [19, 64, 82], "high": [3, 12, 14, 16, 19, 21, 25, 26, 27, 28, 69, 73, 81, 82, 89, 93], "higher": [0, 1, 5, 6, 9, 10, 12, 17, 21, 22, 24, 28, 71, 74, 81, 89, 93, 95], "highest": [6, 7, 22, 23], "highli": [12, 16, 28, 72, 77], "highlight": [22, 25, 77, 79], "himself": 51, "hin": 27, "hint": [73, 82], "histori": 28, "hit": [0, 28, 70, 74, 79, 80, 93], "hk": 12, "ho": 10, "hoc": [19, 87], "hold": [0, 1, 3, 4, 7, 8, 9, 10, 12, 70, 76, 83, 89, 96], "home": [20, 59, 73], "homo_head_pattern": 83, "homogen": 2, "hope": 26, "hopper": [5, 9, 20, 21, 22, 25, 27, 28, 29, 64, 65, 71, 77, 91, 93], "horatio": 51, "horizont": [28, 29], "host": [1, 10, 28, 30, 32, 37, 52, 57, 64, 65, 70, 80, 82, 93], "host_cache_s": 70, "host_context_length": [82, 83, 84, 87, 92], "host_context_progress": [82, 83, 92], "host_cross_kv_cache_block_offset": [83, 87], "host_cross_kv_cache_pool_map": 83, "host_cross_kv_cache_pool_point": 83, "host_kv_cache_block_offset": [82, 83, 87, 92], "host_kv_cache_block_point": 92, "host_kv_cache_pool_map": [82, 83, 92], "host_kv_cache_pool_point": [82, 83, 92], "host_max_attention_window_s": [82, 83, 92], "host_past_key_value_length": [82, 83, 92], "host_request_typ": [82, 83, 84, 92], "host_runtime_perf_knob": [82, 83, 92], "host_sink_token_length": [82, 83, 92], "hostcaches": [0, 9], "hostmemori": 1, "hostnam": 30, "hour": 75, "hous": 76, "how": [0, 2, 3, 12, 14, 16, 17, 19, 29, 32, 40, 55, 64, 69, 70, 72, 75, 77, 78, 80, 82, 88, 89, 90, 92, 94, 96, 97], "howev": [2, 3, 5, 12, 19, 20, 21, 26, 27, 28, 30, 73, 76, 77, 79, 80, 81, 89, 93, 95, 96], "hpc": 22, "html": [1, 82, 92], "http": [0, 1, 4, 10, 19, 20, 26, 29, 30, 33, 34, 35, 60, 61, 62, 65, 66, 67, 69, 82, 88, 90, 92, 93, 94], "hub": [18, 58, 70, 73, 88, 93, 94], "hug": [3, 10, 13, 18, 19, 36, 70, 73, 84, 88, 93], "huggingfac": [0, 10, 14, 15, 17, 19, 20, 34, 58, 61, 69, 73, 74, 75, 88, 91, 92, 93, 95], "huggingface_exampl": 94, "huggingface_hub": 58, "huggingface_model_card": 94, "human": [26, 73], "hurt": [28, 80], "hw": [26, 28], "hybrid": [4, 93], "hyper": 15, "hypothesi": 12, "i": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 23, 24, 25, 26, 27, 28, 29, 30, 32, 33, 35, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 62, 65, 66, 67, 69, 70, 71, 73, 74, 75, 77, 78, 79, 80, 81, 82, 83, 84, 85, 87, 88, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99], "ia3": 5, "iactivationlay": 16, "ibrahimamin1": 93, "ibufferptr": 1, "iconstantlay": 82, "icudaengin": [87, 89], "id": [0, 1, 3, 9, 27, 36, 48, 70, 73, 74, 82, 83, 87, 88, 97, 98], "idea": [10, 28, 80], "ideal": [7, 77, 79, 93], "ident": [3, 9, 28, 29, 82], "identifi": [0, 6, 10, 12, 16, 73, 79, 82], "idl": 0, "idtyp": [0, 3], "idx": 87, "ieee": 90, "ieinsumlay": 82, "ielementwiselay": 82, "iexecutioncontext": [87, 89], "ifb": [12, 93], "ifilllay": 82, "igatherlay": 82, "ignor": [29, 70, 73, 82, 87], "ignore_eo": [70, 93], "igptdecod": 1, "ihostmemori": [1, 16, 87], "ii": [5, 82], "ij": 82, "ijk": 82, "ijl": 82, "ik": 82, "ikl": 82, "ilay": [7, 16], "illustr": [7, 12, 18, 26, 27], "ilogg": 1, "ilooplay": 82, "imag": [30, 34, 39, 55, 56, 57, 61, 64, 66, 67, 73, 83, 87, 93], "image64": 61, "image_grid_thw": 87, "image_patches_indic": 87, "image_path": 87, "image_s": 84, "image_token_index": 87, "image_url": [34, 61], "imatrixmultiplylay": 82, "imbal": 79, "immedi": [5, 12, 71, 75, 92], "immut": 1, "impact": [11, 12, 21, 25, 26, 27, 28, 30, 58, 76, 77, 79, 80, 81], "imped": 25, "impl": [0, 99], "implement": [2, 3, 5, 6, 8, 12, 15, 16, 18, 19, 21, 28, 52, 64, 71, 82, 83, 88, 90, 91, 92, 93, 95, 96, 98, 99], "implicit": [1, 5, 12, 82], "implicitli": 1, "import": [11, 12, 17, 19, 21, 25, 27, 28, 30, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 60, 61, 62, 64, 66, 67, 75, 77, 79, 80, 81, 88, 91, 93, 94, 95, 96, 98], "impos": 25, "improv": [5, 9, 11, 16, 21, 22, 23, 24, 25, 26, 27, 28, 29, 42, 46, 47, 49, 50, 64, 71, 73, 74, 75, 77, 78, 79, 80, 93, 94, 97], "in_channel": 83, "in_featur": [15, 16, 83], "in_hidden_s": 82, "in_len": 7, "in_point": 82, "in_progress": 87, "includ": [0, 1, 2, 3, 5, 6, 9, 10, 12, 15, 16, 17, 18, 21, 22, 24, 27, 28, 29, 30, 36, 45, 51, 59, 65, 67, 69, 70, 71, 77, 80, 82, 88, 90, 92, 93, 96, 97, 98, 99], "include_stop_str_in_output": 70, "inclus": 82, "incompat": [29, 93, 94], "incorpor": [0, 26, 71, 93], "incorrect": [9, 12, 93], "increas": [0, 5, 9, 12, 16, 20, 22, 23, 26, 27, 28, 29, 72, 73, 75, 77, 80, 81, 82, 93, 99], "incred": 71, "increment": [65, 93], "incur": [16, 26], "inde": 89, "independ": [0, 1, 2, 3, 12, 82], "index": [0, 1, 3, 8, 12, 17, 26, 36, 53, 64, 66, 67, 70, 82, 87, 88, 93, 97], "index_select": 82, "indic": [0, 1, 3, 5, 6, 12, 15, 70, 81, 82, 83, 87, 89, 98], "indim": 1, "indimfirst": 1, "indirect": 1, "individu": [26, 93], "indivis": 93, "inductor": 70, "industri": 73, "ineffici": [5, 26], "inetworkdefinit": [7, 16, 82], "inevit": 16, "inf": 52, "infeas": 3, "infer": [0, 2, 6, 10, 12, 16, 18, 19, 20, 21, 22, 23, 26, 28, 29, 34, 61, 64, 69, 72, 74, 75, 76, 77, 78, 80, 81, 82, 87, 90, 92, 93, 96], "infer_shap": 87, "inferencerequest": 93, "infin": 32, "infinit": [16, 73, 74], "inflat": 26, "inflight": [0, 5, 10, 12, 30, 68, 70, 73, 78, 79, 82, 93, 97, 99], "inflight_request_id": 99, "inflightbatch": 0, "inflightbatchingstat": [0, 30], "influenc": [26, 80], "info": [0, 29, 30, 73, 89, 92], "inform": [0, 1, 2, 3, 5, 6, 8, 12, 15, 16, 21, 24, 26, 30, 64, 71, 73, 75, 91, 92, 93], "infti": 6, "inherit": [17, 19, 82, 95, 96, 98, 99], "init": [1, 20, 28, 65, 93], "init_audio_encod": 87, "init_image_encod": 87, "init_llm": 87, "init_processor": 87, "init_token": 87, "initi": [1, 2, 12, 17, 26, 52, 70, 73, 77, 79, 80, 89, 92, 93, 95, 97, 99], "initializer_list": [0, 1], "initmemorypool": 89, "inittozero": 1, "inlin": [0, 1], "inner": 82, "inner_layernorm": [83, 84], "innov": 28, "inp": 82, "inpaint": [34, 61], "inprogress": 1, "input": [0, 1, 3, 6, 7, 9, 10, 11, 12, 16, 17, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 36, 38, 39, 56, 61, 64, 69, 70, 72, 73, 74, 75, 76, 78, 80, 81, 82, 83, 84, 87, 89, 91, 92, 93, 95, 96, 97, 99], "input_1": 82, "input_1_": 82, "input_audio": 87, "input_featur": 84, "input_fil": 93, "input_id": [9, 14, 26, 73, 82, 84, 87, 92, 95], "input_imag": 87, "input_layernorm": [14, 15, 17, 95], "input_length": [82, 83, 84, 87], "input_list": 82, "input_n": 82, "input_n_": 82, "input_text": [14, 16, 87, 88], "input_timing_cach": [29, 70], "input_token_extra_id": 87, "inputbuff": 1, "inputdesc": 16, "inputdtyp": 1, "inputgentokenshost": 1, "inputlen": 1, "inputpack": [1, 6], "inputs_emb": 95, "inputtokenextraid": 0, "inputtokenid": 0, "insert": [7, 16, 73, 82], "insertinputtensor": 1, "insid": [1, 12, 17, 19, 20, 27, 28, 65, 67, 82, 89, 97], "insight": 26, "insiz": 1, "inspect": [29, 72, 89], "inspir": 27, "instabl": 2, "instal": [19, 30, 31, 55, 56, 57, 65, 69, 75, 88, 93, 95], "instanc": [0, 2, 3, 6, 7, 8, 12, 16, 26, 36, 52, 69, 70, 87, 89, 93, 97], "instance_idx": 92, "instanti": [75, 81, 98], "instead": [7, 9, 12, 16, 19, 20, 21, 36, 65, 70, 80, 81, 82, 89, 93], "instruct": [12, 20, 28, 30, 34, 39, 46, 61, 65, 73, 74, 75, 76, 80, 81, 88, 91, 93, 94, 95], "instrument": 28, "int": [0, 1, 6, 14, 15, 16, 19, 48, 52, 70, 79, 82, 83, 84, 87, 95, 97, 98, 99], "int32": [1, 5, 29, 82, 85, 92], "int32_t": [0, 1, 82], "int4": [17, 19, 25, 29, 36, 59, 64, 91, 93], "int4_weight": 90, "int64": [1, 6, 82, 92], "int64_t": [0, 1], "int8": [1, 15, 17, 19, 25, 29, 64, 70, 77, 82, 89, 91, 93], "int8_kv_cach": [5, 90, 93], "int8_t": [0, 1], "int8_weight": 90, "int8awq": 77, "int_clip": 82, "integ": [5, 70, 73, 82, 90, 93], "integr": [12, 93, 96, 97, 98, 99], "intellig": 71, "intend": 89, "intens": 28, "intent": 75, "intention": 19, "intenum": 82, "inter": [2, 75, 76, 77, 79, 80, 92, 93], "inter_layernorm": 84, "inter_s": 17, "interact": [3, 12, 71, 88, 92], "interchang": [8, 69], "interconect": 76, "interconnect": [6, 75, 76, 77, 79, 80], "interest": 73, "interfac": [16, 19, 75, 87, 93, 95, 96], "interfer": 92, "interleav": [5, 16, 28], "intermedi": [5, 16, 28, 70, 92], "intermediate_s": [15, 84], "intern": [1, 3, 5, 8, 19, 20, 26, 28, 75, 78, 89, 92, 98], "internal_error": [29, 30], "internlm": [69, 90, 91, 93], "internlm2": [90, 91, 93], "internvl2": 93, "interpol": 82, "interpolation_scal": 83, "interpret": [3, 65, 79], "intersect": 2, "intertwin": 80, "intflag": [84, 86], "intpsplitdim": 1, "intra": 76, "introduc": [19, 22, 26, 27, 32, 90, 93], "introduct": [78, 88, 93], "intuit": [28, 71, 78], "inv": 82, "inv_freq": 82, "invalid": [92, 93], "invalidateremoteag": 0, "inventori": 73, "invers": 5, "invest": 73, "investig": [20, 93], "invit": 59, "invoc": 93, "invok": [0, 3, 7, 69, 92, 99], "invokequant": 16, "involv": [0, 1, 2, 12, 16, 25, 27, 28, 83, 96, 97, 98], "io": [5, 31, 32, 89, 93], "ip": [0, 93], "ipc": 65, "ipc_uc_handl": 1, "ipc_uc_ptr": 1, "ipc_uc_va": 1, "ipcmemori": 1, "ipcnvl": 1, "ipcnvlsalloc": 1, "ipcnvlsfre": 1, "ipcnvlshandl": 1, "ipcnvlssupport": 1, "ipluginv3lay": 82, "ireducelay": 82, "irrespect": [0, 6, 52, 70], "is_alibi": 82, "is_caus": 83, "is_const_v": 1, "is_cuda_graph": 97, "is_def": 82, "is_dora": 10, "is_dynam": 82, "is_enc_dec": 87, "is_expert": 83, "is_gated_activ": 82, "is_gemma_2": 84, "is_gemma_3": 84, "is_keep_al": 70, "is_loc": 83, "is_medusa_mod": 87, "is_mla_en": 82, "is_mla_enabled_flag": 82, "is_module_excluded_from_quant": 70, "is_mrop": 82, "is_network_input": 82, "is_orchestrator_mod": 87, "is_public_pool": 70, "is_qkv": 83, "is_redrafter_mod": 87, "is_rop": 82, "is_trt_wrapp": 82, "is_use_oldest": 70, "is_valid": 83, "is_valid_cross_attn": 83, "isagentst": 0, "isauto": 0, "isbeamsearch": 0, "iscomplet": 0, "iscontextparallel": 1, "iscontinuouskvcach": 1, "iscrossattent": 1, "isdon": 1, "isdora": 1, "isdrafttokensextern": 1, "iseagl": [0, 1], "iselectlay": 82, "isexplicitdrafttoken": [0, 1], "isexternaldrafttoken": 0, "isfin": [0, 3], "isfirstcontextparallelrank": 1, "isfirstpipelineparallelrank": 1, "isfirsttensorparallelrank": 1, "isgreedysampl": 0, "ishufflelay": 82, "iskvcacheen": 1, "isl": [0, 21, 22, 23, 24, 26, 27, 28, 73, 74, 80], "islastpipelineparallelrank": 1, "isleg": 0, "islicelay": 82, "isload": 1, "islookahead": 0, "islookaheaddecod": 1, "ismedusa": [0, 1], "ismpist": 0, "ismultimod": 1, "isn": 92, "isnon": 1, "isoftmaxlay": 82, "isorchestr": 0, "ispagedkvcach": 1, "isparticip": [0, 93], "ispipelineparallel": 1, "ispoint": 1, "isrnnbas": 1, "issequencefin": [0, 3], "issocketst": 0, "issu": [5, 16, 19, 58, 64, 65, 67, 69, 73, 74, 75, 82, 92], "istensorparallel": 1, "isthreadsaf": 0, "istopk": 0, "istopkandtopp": 0, "istopkortopp": 0, "istopp": 0, "istransformerbas": 1, "istream": [0, 1], "isunsign": 1, "isusebantoken": 0, "isusebanword": 0, "isuseexpliciteosstop": 0, "isusefrequencypenalti": 0, "isusemaxlengthstop": 0, "isuseminlength": 0, "isuseminp": 0, "isusenorepeatngrams": 0, "isuseoccurrencepenalti": 0, "isusepenalti": 0, "isusepresencepenalti": 0, "isuserepetitionpenalti": 0, "isusestopcriteria": 0, "isusestopword": 0, "isusetemperatur": 0, "isusevariablebeamwidthsearch": 0, "iswhisp": 1, "ite": 87, "item": [0, 3, 28, 87], "itensor": [0, 82], "itensorbind": 1, "itensorptr": 1, "iter": [0, 1, 3, 5, 12, 17, 26, 27, 30, 70, 71, 73, 75, 79, 80, 81, 87, 93], "iter_stats_max_iter": 70, "iterationresult": 70, "iterationstat": 0, "iterationtyp": 0, "iterlatencym": [0, 30], "iterlatencymillisec": 93, "iterstat": 0, "iterstatsmaxiter": 0, "iterstatsvec": 0, "ith": 82, "itl": [77, 80, 93], "its": [0, 1, 3, 5, 6, 7, 8, 13, 15, 16, 17, 19, 21, 23, 26, 27, 45, 69, 71, 73, 76, 78, 79, 80, 82, 89, 96, 97, 99], "itself": [3, 28, 87], "itsuji": 73, "iunarylay": 82, "j": [5, 6, 22, 25, 27, 55, 56, 57, 69, 73, 82, 90, 91, 93], "jacobi": 12, "jai": 93, "jamesthez": 93, "jane": 59, "janpetrov": 93, "japanes": [10, 73], "jax": [15, 19], "ji": 82, "jit": [20, 67, 93], "jj": 82, "jk": 82, "jl749": 93, "job": [16, 56, 57], "joint": 28, "joint_attention_kwarg": 84, "joint_attn_forward": 83, "journei": [26, 71], "jpg": 73, "json": [0, 1, 3, 15, 30, 33, 34, 35, 38, 39, 45, 52, 70, 72, 73, 88, 93], "json_object": 70, "jsonconfigstr": 0, "jsonl": 73, "jsonseri": 0, "just": [0, 1, 12, 27, 28, 55, 56, 57, 58, 67, 73, 75, 81, 87, 89], "justic": [42, 46, 47, 49, 50, 58], "k": [1, 5, 6, 10, 12, 18, 26, 27, 28, 70, 82, 90, 92, 93, 95, 97], "k_b_proj_tran": 82, "k_dim": 82, "k_proj": [17, 95], "kattent": 1, "kattn_dens": 1, "kattn_k": 1, "kattn_q": 1, "kattn_qkv": 1, "kattn_v": 1, "kauto": 0, "kbatchedpostprocessornam": [0, 3], "kbeamsearch": 0, "kbf16": 0, "kblk": 0, "kbool": [0, 1], "kbyte_typ": 1, "kc_cache_retention_config": 93, "kcancel": 0, "kchatglm": 1, "kcontext": 1, "kcontext_in_progress": 0, "kcontinu": 1, "kcpu": [0, 1], "kcpu_pin": 0, "kcpu_pinnedpool": 0, "kcross_attn_dens": 1, "kcross_attn_k": 1, "kcross_attn_q": 1, "kcross_attn_qkv": 1, "kcross_attn_v": 1, "kdatatyp": 1, "kdecoder_onli": [0, 13], "kdefault": 0, "kdefault_num_tokens_per_block": 1, "kdefaultbatchsizet": 0, "kdefaultdynamicbatchmovingaveragewindow": 0, "kdefaultgpuspernod": 1, "kdefaultiterstatsmaxiter": 0, "kdefaultlookaheaddecodingngram": 0, "kdefaultlookaheaddecodingverificationset": 0, "kdefaultlookaheaddecodingwindow": 0, "kdefaultmaxadapters": 0, "kdefaultmaxpagesperblockdevic": 0, "kdefaultmaxpagesperblockhost": 0, "kdefaultmaxseqidlemicrosecond": 0, "kdefaultoptimaladapters": 0, "kdefaultprior": 0, "kdefaultrequeststatsmaxiter": 0, "kdefaultretentionprior": 0, "kdisabl": 1, "kdrafttokensextern": 1, "kdram": 0, "kdynamicpostprocessornameprefix": 0, "keagl": [0, 1], "kebnf_grammar": [0, 3], "keep": [0, 5, 6, 19, 26, 28, 70, 74, 81, 82, 93], "keepdim": 82, "kei": [0, 2, 3, 9, 16, 21, 25, 27, 28, 64, 73, 74, 79, 84, 87, 92, 96, 97, 98], "kenabl": 1, "kencdec": 1, "kencoder_decod": 0, "kencoder_in_progress": 0, "kencoder_onli": 0, "kend_id": 0, "kept": [5, 19, 70, 82], "kequal_progress": 0, "kera": 19, "kernel": [1, 5, 9, 16, 21, 27, 28, 29, 52, 67, 71, 72, 77, 80, 82, 87, 88, 89, 92, 93], "kernel_s": [82, 83], "kexplicitdrafttoken": [0, 1], "kexternaldrafttoken": 0, "key_length": [82, 83], "keyvaluecacheparam": [83, 84], "keyword": [17, 70, 82, 89], "kfile": 0, "kfirst_come_first_serv": 0, "kfloat": [1, 16], "kfp16": 0, "kfp32": [0, 70], "kfp8": 0, "kgener": 1, "kgeneration_complet": 0, "kgeneration_in_progress": 0, "kglm": 1, "kgpt": 1, "kgpu": [0, 1], "kguaranteed_no_evict": 0, "khalf": 1, "kind": [4, 5, 7, 26, 99], "kinflight": 0, "king": 51, "kint32": [0, 1], "kint64": [0, 1], "kint8": [0, 1], "kinvalid": 1, "kispoint": 1, "kisunsign": 1, "kj": 82, "kjson": [0, 3], "kjson_schema": [0, 3], "kleader": [0, 2], "klength": 0, "klinear": 1, "klookahead": 0, "klookaheaddecod": 1, "kmamba": 1, "kmax_util": 0, "kmaxretentionprior": 0, "kmedusa": [0, 1], "kminretentionprior": 0, "kmla": 0, "kmlp_4h_to_h": 1, "kmlp_gate": 1, "kmlp_gate_up": 1, "kmlp_h_to_4h": 1, "kmlp_router": 1, "kmoe_4h_to_h": 1, "kmoe_gat": 1, "kmoe_h_to_4h": 1, "kmoe_rout": 1, "kmpi": 0, "knegativeinfin": 1, "knob": [0, 70, 81, 82], "knone": 1, "knoop": 1, "knot_finish": 0, "know": [6, 72, 81, 82], "knowledg": 64, "known": [5, 12, 16, 64, 67, 82, 91], "knumflag": 0, "kobj": 0, "kopt_profiles_split_point": 1, "korchestr": [0, 2], "kosmo": [91, 93], "kpage": 1, "kpin": 1, "kpinnedpool": 1, "kqueu": 0, "kread": 0, "krecurr": 1, "krecurrentgemma": 1, "kregex": [0, 3], "kstatic": 0, "kstatic_batch": 0, "kstop_word": 0, "kstructural_tag": 0, "ktimed_out": 0, "ktopk": 0, "ktopktopp": 0, "ktopp": 0, "ktrtpointertyp": 1, "kuint8": [0, 1], "kunderlyingtyp": 1, "kunish": 10, "kunknown": 0, "kunsign": 1, "kusebantoken": 0, "kusebanword": 0, "kuseexpliciteosstop": 0, "kusefrequencypenalti": 0, "kusemaxlengthstop": 0, "kuseminlength": 0, "kuseminp": 0, "kusenorepeatngrams": 0, "kuseoccurrencepenalti": 0, "kusepenalti": 0, "kusepresencepenalti": 0, "kuserepetitionpenalti": 0, "kusestandardstopcriteria": 0, "kusestopword": 0, "kusetemperatur": 0, "kusevariablebeamwidthsearch": 0, "kuvm": [0, 1], "kv": [0, 1, 2, 3, 10, 16, 19, 21, 25, 27, 28, 29, 30, 36, 40, 41, 43, 44, 54, 64, 68, 70, 71, 73, 74, 75, 79, 82, 87, 88, 93, 94, 95, 96, 97, 99], "kv_b_proj": 82, "kv_cach": 0, "kv_cache_block_offset": [82, 83, 87, 92], "kv_cache_block_point": 92, "kv_cache_config": [30, 36, 43, 44, 49, 51, 53, 54, 70, 81, 98], "kv_cache_dtyp": [51, 70, 73, 77, 86, 98], "kv_cache_enable_block_reus": [87, 93], "kv_cache_free_gpu_mem_fract": [20, 74, 81], "kv_cache_free_gpu_memory_fract": [30, 37, 87, 93], "kv_cache_host_memory_byt": 9, "kv_cache_manag": [0, 93, 96, 97, 98, 99], "kv_cache_param": [83, 84, 97], "kv_cache_quant_algo": [15, 59, 70, 73, 77], "kv_cache_quant_mod": [5, 82], "kv_cache_retention_config": 70, "kv_cache_scaling_factor": [5, 15], "kv_cache_typ": [16, 29, 70, 87, 93], "kv_dtype": 84, "kv_event": 51, "kv_head": 83, "kv_host_cache_byt": 9, "kv_lora_rank": [82, 83], "kv_orig_quant_scal": 82, "kv_quant_orig_scal": 82, "kvalue_status_load": 1, "kvalue_status_miss": 1, "kvalue_status_process": 1, "kvcach": [0, 26, 43, 44, 54, 93], "kvcacheblock": 8, "kvcacheblockpool": 8, "kvcacheconfig": [0, 5, 9, 36, 43, 44, 49, 51, 53, 54, 70, 81, 89], "kvcachecreateddata": [0, 70], "kvcacheev": 0, "kvcacheeventdata": 0, "kvcacheeventdiff": 0, "kvcacheeventmanag": [0, 64], "kvcachehitr": 0, "kvcachehitrateperrequest": 0, "kvcacheindex": 1, "kvcachemanag": [0, 5, 9, 87, 97, 98], "kvcachemetr": 0, "kvcacheparam": 97, "kvcacheremoveddata": [0, 70], "kvcacheretentionconfig": [0, 70], "kvcaches": 0, "kvcachestat": [0, 30], "kvcachestoredblockdata": 0, "kvcachestoreddata": [0, 70], "kvcachetransferend": 0, "kvcachetransferm": 0, "kvcachetransfermod": [0, 70], "kvcachetransferstart": 0, "kvcachetyp": [1, 70, 87], "kvcachetypefromstr": 1, "kvcacheupdateddata": [0, 70], "kvfactor": 0, "kvheadnum": 82, "kvram": 0, "kwarg": [17, 19, 70, 82, 83, 84, 87, 93, 95], "kwrite": 0, "kxgrammar": 0, "l": [12, 55, 56, 57, 73, 91], "l2": 29, "l20": 29, "l304": 26, "l345": 26, "l4": 29, "l40": 29, "l440": 26, "l506": 26, "l546": 26, "l823": 26, "lab": 73, "label": [7, 82, 83, 84], "labelembed": 83, "lack": 0, "lai": 27, "lambda": [0, 3], "lamportinitializeal": 1, "languag": [0, 6, 12, 16, 18, 21, 26, 71, 72, 82, 90, 91, 93, 96], "language_adapt": [87, 93], "language_adapter_config": 87, "language_adapter_rout": [84, 87], "language_adapter_uid": 87, "language_model": 17, "languageadapterconfig": 87, "languageadapteruid": 0, "larg": [5, 9, 11, 12, 16, 18, 19, 20, 21, 25, 26, 28, 29, 34, 52, 61, 71, 72, 73, 76, 77, 79, 80, 82, 89, 91, 92, 93, 96], "larger": [0, 2, 5, 6, 9, 12, 13, 20, 22, 23, 25, 28, 54, 70, 73, 74, 82, 87, 89, 93], "largest": [6, 21, 22, 23, 82], "last": [0, 1, 3, 5, 10, 11, 12, 14, 26, 27, 70, 79, 81, 82, 84], "last_lay": 87, "last_process_for_ub": 82, "last_token_id": [82, 84, 92], "last_token_ids_for_logit": 84, "last_tokens_id": 82, "lastdraftindic": 1, "lastdraftlen": 1, "lastdraftpath": 1, "lastdrafttoken": 1, "lastgenerationlength": 1, "lastit": 0, "lastpositionidsbas": 1, "lasttokentim": 0, "late": 58, "latenc": [0, 5, 9, 12, 22, 23, 25, 27, 28, 29, 64, 70, 74, 79, 80, 81, 82, 93], "latent": [28, 83, 84], "later": [0, 1, 6, 10, 12, 16, 19, 23, 46, 49, 69, 77, 80, 87, 89, 92, 94], "latest": [0, 28, 31, 65, 88, 93], "latter": [3, 25, 93], "launch": [2, 9, 16, 28, 30, 52, 55, 56, 57, 64, 67, 69, 75, 92, 93, 94], "launch_llama_3": 16, "layer": [0, 1, 2, 4, 5, 6, 7, 8, 10, 12, 14, 15, 16, 17, 27, 29, 70, 76, 82, 87, 88, 89, 90, 92, 93, 95, 97, 98], "layer1": 10, "layer_idx": [10, 14, 82, 87, 95, 97], "layer_names_onli": [29, 70], "layer_norm": [82, 83], "layer_quant_mod": 70, "layer_typ": 87, "layerid": [1, 10], "layeridx": 1, "layernorm": [14, 29, 80, 82, 83, 93], "layernorm_shar": 83, "layernorm_typ": 83, "layernormpositiontyp": 82, "layernormtyp": [82, 83], "layertyp": [1, 7], "layerwis": 70, "layout": [79, 93], "lead": [7, 9, 12, 16, 29, 58, 65, 73, 74, 75, 77, 79, 80], "leader": [0, 87], "learn": [22, 23, 25, 42, 46, 47, 49, 50, 52, 77, 82, 88], "learned_absolut": [15, 82, 83, 84], "least": [0, 3, 5, 19, 20, 30, 58, 79, 87], "leav": [59, 79, 80, 81], "left": [70, 74, 79, 81, 82], "legaci": [17, 81, 85, 93], "len": [0, 1, 73, 82, 87, 99], "length": [0, 1, 5, 9, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 53, 70, 73, 74, 75, 78, 80, 81, 82, 87, 89, 92, 93, 97, 98], "length_penalti": [6, 70, 87], "lengthlengthpenalti": 6, "lengthpenalti": [0, 1, 6], "less": [0, 3, 5, 6, 16, 22, 27, 70, 74, 82], "let": [7, 14, 15, 17, 26, 31, 36, 71, 73, 79, 82], "letter": 82, "level": [0, 1, 3, 5, 8, 10, 14, 15, 17, 19, 27, 28, 29, 30, 49, 69, 72, 73, 89, 93, 95], "leverag": [12, 21, 26, 27, 77, 88], "lf": [10, 20, 65, 69], "lfz941": 93, "lh": 1, "lib": [19, 67, 73], "libnam": 0, "libnvinfer_plugin_tensorrt_llm": 65, "libopenmpi": [66, 67], "librari": [16, 18, 65, 69, 71, 92, 93, 97], "libtensorrt_llm": 65, "licens": [69, 88], "life": 58, "lifecycl": 8, "lightweight": 5, "like": [0, 3, 5, 6, 7, 9, 12, 15, 16, 18, 19, 25, 26, 27, 28, 29, 36, 42, 45, 46, 47, 48, 49, 50, 51, 52, 54, 55, 56, 57, 58, 59, 70, 71, 73, 75, 76, 77, 79, 80, 81, 82, 88, 89, 90, 92, 93, 94, 95, 96, 98], "likelihood": [4, 9, 12], "limit": [0, 2, 3, 5, 6, 7, 16, 19, 20, 25, 26, 27, 28, 36, 67, 69, 70, 71, 75, 79, 81, 82, 85, 87, 89, 91, 97], "lin": 21, "line": [9, 20, 25, 73, 75, 77, 80, 89, 93, 98, 99], "linear": [1, 10, 12, 14, 15, 16, 28, 82, 89, 90, 93, 95, 97], "linearactiv": 83, "linearapproximategelu": 83, "linearbas": 83, "lineargeglu": 83, "lineargelu": 83, "linearli": 89, "linearswiglu": 83, "link": [9, 20, 26, 31, 32, 93], "linspac": 82, "linux": [64, 91, 93], "linux_x86_64": 65, "list": [0, 1, 3, 5, 6, 7, 15, 16, 17, 18, 27, 36, 52, 65, 68, 70, 71, 73, 74, 75, 82, 83, 84, 87, 91, 92, 93, 97, 98, 99], "list_siz": 83, "liter": 70, "littl": [27, 80], "live": 89, "livecodebench": 26, "lkm2835": 93, "ll": [25, 30], "llama": [6, 10, 12, 13, 17, 19, 22, 23, 25, 29, 46, 54, 69, 75, 76, 78, 79, 81, 88, 90, 91, 93, 94, 95], "llama2": [5, 10, 21, 22, 93], "llama3": 82, "llama4": 70, "llama4forconditionalgener": 91, "llama_13b": 23, "llama_70b": 23, "llama_7b": [10, 13], "llama_7b_with_lora_qkv": 10, "llama_model_path": 36, "llamaconfig": [84, 95], "llamaforcausallm": [17, 19, 84, 91], "llamamodel": 84, "llava": [17, 90, 91, 93], "llava_dict": 17, "llavallamamodel": 91, "llavanextforconditionalgener": 91, "llavanextvisionconfig": 84, "llavanextvisionwrapp": 84, "llm": [0, 2, 3, 5, 6, 7, 8, 9, 10, 11, 14, 16, 21, 24, 26, 28, 29, 30, 33, 34, 35, 37, 38, 39, 43, 44, 45, 46, 47, 48, 49, 51, 52, 53, 54, 58, 59, 60, 61, 62, 66, 67, 68, 70, 72, 74, 76, 77, 78, 80, 81, 82, 84, 86, 87, 90, 92, 94, 95, 96, 97, 98, 99], "llm_arg": [70, 74], "llm_engine_dir": 87, "llm_inference_distribut": 69, "llm_kwarg": [43, 44, 54], "llm_mgmn_": 93, "llm_option": 74, "llm_ptq": 94, "llmapi": [3, 30, 36, 43, 44, 45, 49, 51, 53, 54, 55, 56, 57, 59, 70, 74, 77, 93], "llmarg": [70, 74, 93], "llmrequest": [1, 98, 99], "llmrequestptr": 1, "llmrequestst": 99, "lm": 12, "lm_head": [14, 17, 54, 73, 93], "lmm": [6, 73], "lmsy": [43, 44, 54], "ln_emb": 17, "ln_f": [14, 17], "load": [0, 1, 10, 14, 15, 16, 19, 24, 26, 28, 29, 46, 49, 54, 67, 69, 70, 73, 74, 75, 80, 81, 84, 86, 87, 88, 89, 93], "load_format": 70, "load_model_on_cpu": 84, "load_tensor": 17, "load_test_audio": 87, "load_test_data": 87, "load_weight": 95, "loaded_weight": 83, "loader": 93, "loadformat": 70, "loadinprogress": 1, "loadremoteag": 0, "loadweight": 1, "local": [15, 16, 20, 26, 29, 46, 47, 48, 49, 50, 55, 56, 57, 59, 65, 67, 70, 73, 74, 77, 93, 98], "local_in_featur": 83, "local_layer_idx": 83, "local_model": [55, 56, 57], "local_out_featur": 83, "local_us": [20, 65, 88], "localhost": [30, 33, 34, 35, 37, 38, 39, 60, 61, 62, 88], "localinadapters": 1, "localindim": 1, "localinouts": 1, "localins": 1, "localoutadapters": 1, "localoutdim": 1, "localouts": 1, "localreduct": 26, "localscaless": 1, "localtotals": 1, "locat": [6, 7, 16, 28, 59, 65, 73, 74, 82, 88, 92, 97], "locate_accepted_draft_token": 87, "lock": [67, 73], "lockstep": 0, "log": [0, 1, 5, 8, 29, 30, 31, 55, 56, 57, 59, 70, 73, 82, 88, 89, 93], "log_level": [29, 30], "log_softmax": 82, "logic": [3, 8, 17, 19, 52, 83, 93, 95, 96, 99], "login": [31, 88], "logit": [0, 1, 6, 12, 26, 27, 40, 41, 70, 73, 82, 84, 87, 92, 93], "logits_dtyp": [15, 29, 84], "logits_processor": [52, 70, 87], "logits_processor_map": 87, "logits_processor_nam": 87, "logitspostprocessor": 0, "logitspostprocessorbatch": [0, 3], "logitspostprocessorconfig": [0, 3, 93], "logitspostprocessormap": 0, "logitspostprocessornam": 0, "logitsprocessor": [52, 70, 87, 93], "logitsprocessorlist": 87, "logitsvec": 1, "logn": [82, 93], "logn_scal": 82, "logprob": [0, 1, 36, 53, 70, 88], "logprobs_diff": 70, "logprobscba": 1, "logprobstil": 1, "london": 92, "long": [5, 25, 29, 72, 73, 75, 76, 77, 79, 80, 89, 93], "long_mscal": [82, 83], "long_rop": 82, "long_rope_embed_posit": 83, "long_rope_embed_positions_for_gpt_attent": 83, "long_rope_rotary_cos_sin": 82, "long_rope_rotary_inv_freq": [82, 83], "longer": [0, 6, 9, 26, 28, 70, 74, 79, 82, 99], "longest": [2, 27, 79, 82], "longrop": 82, "longtensor": [87, 95], "look": [0, 3, 19, 24, 65, 71, 73, 93], "lookahead": [0, 1, 40, 41, 64, 70, 93], "lookahead_config": [53, 70, 87], "lookahead_decod": [29, 84], "lookaheadconfig": 0, "lookaheaddecod": 1, "lookaheaddecodingbuff": 1, "lookaheaddecodingconfig": [0, 1, 53, 70], "lookaheadinput": 1, "lookaheadoutput": 1, "lookaheadruntimebuff": 1, "lookaheadruntimeconfig": 1, "lookup": [64, 82, 83, 93], "lookup_plugin": 82, "loop": [0, 3, 6, 16, 17, 70, 81], "lopuhin": 93, "lora": [0, 1, 3, 40, 41, 64, 68, 70, 82, 83, 84, 87, 93], "lora_ckpt_sourc": [29, 87], "lora_config": [58, 70, 84], "lora_dir": [10, 29, 58, 87], "lora_dir1": 58, "lora_dir2": 58, "lora_dir3": 58, "lora_hidden_st": 83, "lora_layer_param": 83, "lora_manag": [58, 70, 87, 93], "lora_param": 84, "lora_plugin": [10, 29, 82, 87], "lora_rank": [10, 82], "lora_request": [58, 70], "lora_runtime_param": 83, "lora_target_modul": [10, 29, 84, 87], "lora_task_uid": 87, "lora_uid": 87, "lora_weights_point": 82, "loracachefullexcept": 1, "loracachepagemanag": 1, "loraconfig": [0, 10, 58, 70, 84, 93], "loraexpectedexcept": 1, "loraid": 0, "loramanag": 87, "loramodulenam": 1, "loraparam": 84, "loraprefetchdir": 0, "lorarequest": [58, 70], "loraruntimeparam": 83, "lorataskidtyp": [0, 1], "loraweight": 10, "loss": [25, 77], "lot": [5, 9, 16, 18, 27], "loudspeak": 23, "lovelac": [71, 91, 93], "low": [5, 14, 19, 20, 25, 26, 27, 28, 29, 64, 82, 93], "low_latency_gemm": 82, "low_latency_gemm_plugin": [29, 73, 77, 83], "low_latency_gemm_swiglu": 82, "low_latency_gemm_swiglu_plugin": [29, 77, 85], "low_rank": 82, "lower": [0, 1, 2, 6, 7, 9, 10, 24, 25, 28, 49, 70, 74, 77, 82, 89], "lowprecis": [11, 82], "lru": [1, 9, 82], "lt": 82, "luotuo": 10, "m": [0, 20, 22, 26, 30, 38, 39, 45, 58, 73, 74, 75, 77, 79, 80, 82, 89, 90], "macceptancethreshold": 0, "machin": [9, 20, 25, 52, 93], "madditionalmodeloutput": 0, "maddr": 0, "made": [52, 71, 93, 99], "magentnam": 0, "mahmoudashraf97": 93, "mai": [0, 1, 2, 3, 5, 9, 10, 11, 12, 15, 16, 17, 19, 20, 26, 27, 29, 31, 55, 56, 57, 65, 67, 69, 72, 73, 74, 75, 80, 81, 82, 83, 85, 89, 92, 93, 95, 96, 97, 98], "main": [3, 6, 8, 21, 24, 26, 27, 34, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 58, 59, 61, 66, 67, 69, 70, 72, 75, 77, 80, 81, 82, 88, 89, 92, 94, 95], "mainli": 27, "maintain": [2, 10, 21, 22, 25, 73, 77, 90], "major": [19, 26, 59, 71, 74, 89], "make": [1, 2, 5, 7, 10, 16, 19, 20, 25, 26, 27, 31, 32, 53, 58, 64, 65, 71, 73, 75, 81, 82, 88, 92, 93], "make_causal_mask": 83, "makeshap": 1, "maketransferag": 0, "mallotedtim": 0, "mallreducecommptr": 1, "mamba": [29, 69, 82, 90, 91, 93], "mamba1": 82, "mamba2": [82, 93], "mamba_conv1d": 82, "mamba_conv1d_plugin": [29, 87], "mamba_vers": 82, "mambaconfig": 84, "mambaforcausallm": 84, "manag": [0, 1, 2, 5, 12, 16, 28, 29, 36, 64, 67, 69, 75, 81, 85, 87, 88, 89, 93, 94, 96, 97], "managedweight": 0, "managedweightsmap": 1, "manageweightstyp": 1, "manageweighttyp": 1, "mandatori": [1, 3, 15], "mani": [0, 5, 8, 9, 12, 16, 19, 27, 28, 29, 32, 59, 70, 74, 77, 79, 81, 82, 91, 92], "manipul": 7, "manner": 7, "mantissa": 22, "manual": [28, 36, 70, 87, 92], "manufactur": 73, "map": [0, 1, 2, 3, 5, 7, 11, 14, 15, 16, 17, 19, 26, 74, 82, 83, 84, 87, 88, 98], "marcellu": 51, "mard1no": 93, "margin": [73, 79], "mark": [1, 7, 79, 82, 92], "mark_as_remov": 7, "mark_output": [3, 82], "markalldon": 1, "markdon": 1, "marker": 70, "marks101": 93, "marktaskdon": 1, "mask": [0, 1, 5, 12, 26, 27, 52, 82, 83, 84, 87, 97], "mask_typ": 82, "masked_scatt": 82, "masked_scatter_": 82, "masked_select": [82, 93], "massiv": 20, "master": [76, 77, 78], "mat2": 82, "match": [0, 4, 7, 12, 27, 64, 70, 73, 82, 83, 87, 88, 92, 93], "match_and_rewrit": 7, "materi": 3, "math": [26, 28, 91], "matichon": 93, "matmul": [5, 16, 29, 77, 82, 90], "matric": 4, "matrix": [5, 16, 24, 28, 64, 71, 73, 76, 82, 88, 97], "mattentionconfig": 0, "mattentiontyp": 0, "matter": 9, "matur": 30, "max": [0, 1, 10, 21, 22, 23, 28, 64, 70, 75, 77, 78, 80, 82, 87, 89, 92, 97], "max_all_reduce_block": 1, "max_attention_window": [70, 81, 93], "max_attention_window_s": [5, 81, 82, 87], "max_attn_valu": 83, "max_batch_s": [5, 10, 13, 15, 16, 19, 20, 27, 29, 30, 36, 37, 43, 44, 49, 53, 54, 70, 73, 77, 79, 80, 82, 84, 87, 89, 92, 93, 98], "max_beam_width": [3, 5, 29, 30, 36, 49, 70, 82, 84, 87, 89], "max_block": [82, 99], "max_blocks_per_seq": 87, "max_blocks_per_sequ": 82, "max_boost_slid": 73, "max_cache_storage_gb": 70, "max_context_length": [82, 83, 87, 89], "max_cpu_lora": 70, "max_decoder_input_len": 84, "max_decoder_seq_len": 29, "max_dist": [5, 82, 83], "max_draft_len": [29, 43, 44, 54, 70, 84, 86], "max_draft_token": [84, 87], "max_encoder_input_len": [29, 70, 84], "max_gen_token": 84, "max_input_len": [10, 13, 15, 16, 29, 70, 73, 84, 87, 89], "max_input_length": [82, 83, 84, 87], "max_kv_seqlen": 82, "max_lora": 70, "max_lora_rank": [10, 29, 58, 70], "max_low_rank": 82, "max_matching_ngram_s": 70, "max_medusa_token": 87, "max_multimodal_len": 29, "max_new_token": [87, 89], "max_ngram_s": [53, 70], "max_non_leaves_per_lay": [43, 44, 70], "max_num_request": [97, 98, 99], "max_num_token": [20, 29, 30, 36, 37, 49, 70, 73, 77, 79, 80, 84, 89, 93, 97], "max_output_len": [16, 87, 88, 92, 93], "max_period": 83, "max_position_embed": [15, 82, 83, 84], "max_position_embedding_len": 82, "max_power_limit": 73, "max_prompt_adapter_token": 70, "max_prompt_embedding_table_s": [29, 70, 87, 93], "max_record": 70, "max_seq_len": [10, 13, 15, 16, 29, 30, 43, 44, 54, 70, 73, 81, 82, 83, 84, 87, 89, 93, 98], "max_seqlen": [5, 82], "max_seqlen_for_logn_sc": 83, "max_sequence_length": [5, 87], "max_token": [30, 33, 34, 35, 45, 51, 60, 61, 62, 70, 81, 88, 94], "max_tokens_in_paged_kv_cach": [81, 87, 93], "max_util": [0, 70, 81], "max_verification_set_s": [53, 70], "max_window_s": [53, 70], "maxaccepteddrafttokensperstep": 1, "maxacceptedtoken": 1, "maxadapters": 0, "maxattentionwindow": 1, "maxattentionwindowvec": [0, 1], "maxbadwordslen": 1, "maxbatchs": [0, 1, 6], "maxbatchsizeruntim": 0, "maxbatchsizeruntimeupperbound": 0, "maxbatchsizestat": 0, "maxbatchsizetunerrecommend": 0, "maxbeamwidth": [0, 1, 3, 93], "maxdecoderstep": 1, "maxdecodingdrafttoken": 1, "maxdecodingtoken": [0, 1], "maxdraftpathlen": [0, 1], "maxdrafttoken": [0, 1], "maxencoderlen": 1, "maxgenerationlength": 1, "maxgenlengthdevic": 1, "maxgenlengthhost": 1, "maxgentoken": 1, "maxim": [0, 21, 23, 26, 28, 73, 81], "maximum": [0, 1, 2, 3, 5, 6, 20, 23, 29, 30, 70, 73, 74, 77, 82, 83, 87, 89, 92, 93, 98], "maxinputlen": [1, 6], "maxinputlength": 1, "maxlength": 1, "maxlengthstop": 0, "maxlorarank": 1, "maxmedusahead": 1, "maxnewtoken": [1, 93], "maxnonleafnodesperlay": 1, "maxnumactiverequest": 0, "maxnumblock": 0, "maxnumpath": 1, "maxnumsequ": [1, 93], "maxnumtoken": [0, 1], "maxnumtokensruntim": 0, "maxnumtokensstat": 0, "maxnumtokenstunerrecommend": 0, "maxoutputlength": 3, "maxpagesperblock": 1, "maxpagesperblockdevic": 0, "maxpagesperblockhost": 0, "maxpathdraftlen": 1, "maxpathlen": [0, 1], "maxpositionembed": [0, 1], "maxpromptembeddingtables": 1, "maxqueues": 0, "maxseqidlemicrosecond": 0, "maxseqlen": 1, "maxsequencelen": [1, 6], "maxsequencelength": 1, "maxstopwordslen": 1, "maxtoken": [0, 89, 93], "maxtokensperenginestep": 1, "maxtokensperstep": 1, "mb": [70, 89], "mbackend": 0, "mbackendagentdesc": 0, "mbart": [91, 93], "mbatchingtyp": 0, "mbatchsizet": 0, "mbeamsearchbuff": 1, "mbeamsearchdiversityr": 0, "mbeamwidth": 0, "mbeamwidtharrai": 0, "mbp": 45, "mbuffer": 1, "mbuffermanag": 1, "mc_handl": 1, "mc_ptr": 1, "mc_va": 1, "mcachemap": 1, "mcachemutex": 1, "mcachepagemanag": 1, "mcachest": 0, "mcachetransceiverconfig": 0, "mcapacityschedulerpolici": 0, "mcommmod": 0, "mcommptr": 1, "mcommstat": 0, "mcommtyp": 0, "mcomputecontextlogit": 1, "mcomputegenerationlogit": 1, "mconfig": [0, 1], "mconnectioninfo": 0, "mcontextchunkingpolici": 0, "mcontextfmha": 1, "mcontextparallel": 1, "mcopyonpartialreus": 0, "mcpu": 1, "mcpudiff": 1, "mcrosskvcachefract": 0, "mcudagraphcaches": 0, "mcudagraphmod": 0, "mcumlogprobstmp": 1, "md": [2, 12, 14, 26, 82, 93, 96], "mdatatyp": [0, 1], "mdebugconfig": 0, "mdebuginputtensor": 0, "mdebugoutputtensor": 0, "mdebugtensornam": 0, "mdebugtensorsmaxiter": 0, "mdecod": 1, "mdecodedurationm": 0, "mdecoderetentionprior": 0, "mdecoderst": 1, "mdecoderstream": 1, "mdecodingconfig": 0, "mdecodinglayerworkspac": 1, "mdecodingmod": [0, 1], "mdefaulteaglechoic": 1, "mdefaultmedusachoic": 1, "mdefaultposteriorthreshold": 1, "mdesc": 0, "mdevic": 1, "mdevicebuffermanag": 1, "mdevicecacheperc": 0, "mdeviceid": [0, 1], "mdirectori": 0, "mdllmutex": 0, "mdogreedysampl": 1, "mdonetask": 1, "mdprank": 0, "mdpsize": 0, "mdrafttoken": 0, "mdstdesc": 0, "mdynamicbatchconfig": 0, "mdynamicbatchmovingaveragewindow": 0, "mdynamicdecodelay": 1, "mdynamictreemaxtopk": 0, "me": [34, 58, 59, 61, 88], "meaglechoic": 0, "meagleconfig": 0, "mean": [1, 4, 5, 6, 9, 12, 15, 17, 19, 20, 22, 23, 27, 28, 30, 38, 39, 56, 58, 70, 72, 73, 74, 75, 76, 81, 82, 85, 87, 89], "meaning": [1, 28, 77, 80], "meant": 78, "mearlystop": 0, "measur": [0, 21, 23, 24, 25, 27, 28, 64, 73, 75, 93], "mechan": [3, 16, 98, 99], "media": [73, 93], "media_path": 73, "medium": [25, 92, 93], "medusa": [0, 1, 29, 40, 41, 64, 70, 82, 84, 87, 93], "medusa_choic": [12, 54, 70, 73, 87], "medusa_decode_and_verifi": 87, "medusa_hidden_act": 86, "medusa_logit": 87, "medusa_model_dir": 86, "medusa_output_token": 87, "medusa_path": 87, "medusa_position_offset": 87, "medusa_temperatur": [12, 87], "medusa_topk": 87, "medusa_tree_id": 87, "medusachoic": [0, 1], "medusaconfig": 84, "medusacurtokensperstep": 1, "medusadecodingconfig": [54, 70], "medusaforcausallm": 84, "medusainput": 1, "medusalogit": 1, "medusapath": 1, "medusatargettokensperstep": 1, "medusatreeid": 1, "meet": [25, 82], "membeddingt": 0, "member": [0, 1, 6, 7, 13, 16, 59, 82], "memlock": [65, 92], "memori": [0, 1, 2, 4, 5, 6, 8, 10, 16, 17, 19, 21, 22, 24, 25, 26, 27, 28, 29, 30, 36, 52, 64, 70, 73, 74, 75, 79, 80, 82, 87, 92, 93, 97, 98], "memorydesc": 0, "memorypoolfre": [1, 89], "memorypoolreserv": [1, 89], "memorypooltrimto": 1, "memorypoolus": 1, "memorytyp": [0, 1], "memorytypestr": 1, "memtyp": 1, "memusagechang": 89, "menableattentiondp": [0, 1], "menablebatchsizetun": 0, "menableblockreus": 0, "menablechunkedcontext": 0, "menablecontextfmhafp32acc": 0, "menablemaxnumtokenstun": 0, "menablepartialreus": 0, "menabletrtoverlap": 0, "mencodedvocab": 0, "mencoderhiddens": 1, "mengineaddr": 1, "menginebuff": 1, "menginepath": 1, "mengines": 1, "mental": 58, "mention": [6, 19, 20, 36, 77], "menu": [31, 32], "merg": [26, 82], "meshgrid": 82, "meshgrid2d": 82, "messag": [11, 26, 30, 33, 34, 60, 61, 67, 70, 74, 82, 88, 89, 93], "met": [0, 1, 3, 12], "meta": [19, 69, 70, 73, 74, 75, 81, 88, 91], "meta_ckpt_dir": 84, "metadata": [8, 95, 97], "metal": [93, 94], "meth": 69, "method": [0, 1, 3, 5, 6, 12, 13, 15, 16, 19, 21, 27, 28, 36, 52, 67, 70, 73, 87, 90, 92, 93, 95, 96, 98, 99], "metric": [0, 28, 70, 72, 73, 74, 75, 77, 79, 80, 93], "mevent": 1, "meventbuffermaxs": 0, "mexecutionconfig": 1, "mextendedruntimeperfknobconfig": 0, "mfastlogit": 0, "mfinishedstep": 1, "mfirstgentoken": 0, "mflagptr": 1, "mfreegpumemoryfract": 0, "mfreepageid": 1, "mfrequencypenalti": 0, "mfuntowicz": 93, "mgathergenerationlogit": 0, "mgemmallreducedtyp": 1, "mgmn": [40, 41], "mgpu": 1, "mgpudiff": 1, "mgpuspernod": 1, "mgpuweightsperc": 0, "mgreedysampl": 0, "mguid": 0, "mguideddecodingconfig": 0, "mguidetyp": 0, "mh": 12, "mh1": 12, "mha": [5, 8, 21, 28, 29, 82, 87, 97], "mhandler": 0, "mhiddens": 1, "mhostcaches": 0, "mi": 90, "mib": 89, "micro": [0, 89], "microbatchid": 0, "microbatchschedul": [96, 99], "microsecond": 0, "microsoft": 15, "middl": 72, "might": [0, 3, 16, 19, 20, 25, 29, 65, 69, 71, 73, 75, 76, 80, 87, 89, 92, 93, 98], "migrat": [19, 85, 93], "million": [59, 73], "millisecond": 0, "millisecondstyp": 0, "mimpl": 0, "min": [0, 1, 6, 22, 26, 27, 28, 70, 73, 75, 80, 82, 92], "min_lat": 82, "min_length": [6, 87], "min_p": [0, 6, 70, 87], "min_token": 70, "mind": [25, 81], "mindim": 1, "mindimfirst": 1, "mini": 93, "minim": [26, 79, 88], "minimum": [0, 5, 6, 70, 73, 74, 77, 82, 87, 89], "minitron": [91, 93], "minittozero": 1, "minlength": [1, 6, 93], "minnormedscorescba": 1, "minor": [59, 93], "minp": [0, 1, 6], "minprogresstask": 1, "minputpack": 1, "minputtokenextraid": 0, "mintoken": [0, 93], "mintpsplitdim": 1, "minut": [0, 25, 75], "mip": 0, "mipcmemoryhandl": 1, "mirco": 0, "mish": 83, "mismatch": [19, 67, 92], "misorchestr": 0, "mispagefre": 1, "miss": [0, 7, 20, 73, 93], "missedblock": 0, "missedblocksperrequest": 0, "mission": 26, "mistral": [4, 69, 73, 77, 80, 90, 91, 93], "mistralai": [73, 91], "mistralforcausallm": 91, "misus": 93, "miterstatsmaxiter": 0, "mitig": [19, 26], "mix": [2, 28, 76, 80, 93], "mixed_precis": 70, "mixed_sampl": 70, "mixer": 93, "mixtral": [4, 10, 69, 73, 77, 80, 90, 91, 93], "mixtralforcausallm": 91, "mixtur": [28, 64, 80, 93], "mjointdecodinginput": 1, "mjointdecodingoutput": 1, "mkdir": 31, "mkdtemp": [46, 49], "mkvcacheconfig": 0, "mkvcachetyp": 1, "mkvfactor": 0, "ml": [82, 93], "mla": [26, 27, 82, 93], "mlayertyp": 1, "mlen": 0, "mlengthpenalti": 0, "mllama": [91, 93], "mllamaconfig": 84, "mllamaforcausallm": 84, "mllamaforconditionalgener": 91, "mlogit": 0, "mlogitsdtyp": 1, "mlogitspostprocessorconfig": 0, "mlookaheaddecodingconfig": 0, "mlookaheaddecodingmaxnumrequest": 0, "mloramodul": 1, "mloraprefetchdir": 0, "mlp": [10, 14, 16, 17, 29, 82, 92, 93, 95], "mlp_4h_to_h": [10, 29], "mlp_bia": 84, "mlp_gate": [10, 29], "mlp_gate_up": [10, 29], "mlp_h_to_4h": [10, 29], "mlp_output": 92, "mlp_router": [10, 29], "mlphiddens": 1, "mlptype": 82, "mm": 93, "mm_data": 73, "mm_embedding_offload": 87, "mma": [28, 82], "mmanag": 1, "mmanagedweightsmap": 1, "mmanageweightstyp": 1, "mmaxadapters": 0, "mmaxattentionwindow": 0, "mmaxattentionwindowvec": 0, "mmaxbatchs": [0, 1], "mmaxbeamwidth": [0, 1], "mmaxdecodingdecodertoken": 1, "mmaxdecodingdrafttoken": 1, "mmaxdecodingenginetoken": 1, "mmaxdraftpathlen": 1, "mmaxencoderlen": 1, "mmaxinputlen": 1, "mmaxlorarank": 1, "mmaxnonleafnodesperlay": 1, "mmaxnumpackedmask": 1, "mmaxnumpath": 1, "mmaxnumtoken": [0, 1], "mmaxpagesperblock": 1, "mmaxpagesperblockdevic": 0, "mmaxpagesperblockhost": 0, "mmaxpositionembed": 1, "mmaxpromptembeddingtables": 1, "mmaxqueues": 0, "mmaxseqidlemicrosecond": 0, "mmaxsequencelen": 1, "mmaxsequencelength": 1, "mmaxtoken": 0, "mmedusachoic": 0, "mmemorytyp": 1, "mmha": [82, 93], "mminp": 0, "mmintoken": 0, "mmlphiddens": 1, "mmlu": [25, 26, 93], "mmlu_llmapi": 93, "mmmu": 73, "mmodelconfig": [0, 1], "mmodelnam": 1, "mmodelvari": 1, "mmoduleidtomodul": 1, "mmropepositiondelta": 0, "mmroperotarycossin": 0, "mmultiblockmod": 0, "mname": [0, 1], "mnbattentionlay": 1, "mnbhead": 1, "mnbkvheadsperlay": 0, "mnblayer": 1, "mnbrnnlayer": 1, "mngramsiz": 0, "mnorepeatngrams": 0, "mnormalizelogprob": 0, "mnumcopystream": [0, 1], "mnumdecodingenginetoken": 1, "mnumdevicemodulelay": 0, "mnumensurework": 0, "mnumhostmodulelay": 0, "mnumkvheadsperattentionlay": 1, "mnumkvheadspercrossattentionlay": 1, "mnumlanguag": 1, "mnumnod": 0, "mnumputwork": 0, "mnumreturnbeam": 0, "mnumreturnsequ": 0, "mnumsm": 1, "mnumtransformerslay": 1, "modal": 90, "mode": [0, 1, 4, 5, 7, 16, 17, 28, 29, 30, 45, 55, 56, 57, 70, 81, 82, 83, 87, 89, 90, 93, 95], "model": [0, 1, 2, 3, 4, 5, 8, 9, 10, 11, 13, 15, 19, 21, 22, 23, 24, 25, 28, 29, 30, 33, 34, 35, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 64, 66, 67, 70, 71, 72, 75, 78, 81, 82, 83, 85, 86, 87, 89, 90, 94, 97, 98, 99], "model_architectur": 70, "model_cl": 83, "model_cls_fil": 29, "model_cls_nam": 29, "model_config": [29, 70, 87, 95], "model_construct": 70, "model_dir": [10, 13, 14, 15, 16, 17, 19, 27, 54, 55, 73, 76, 84, 86, 88, 92], "model_engin": 98, "model_nam": [56, 74, 87], "model_path": [13, 56, 72, 73], "model_post_init": 70, "model_qu": 73, "model_weights_load": [17, 93], "modelconfig": [0, 6, 87, 93, 95], "modelengin": [96, 98], "modelidtomodel": 1, "modeling_deepseekv3": [26, 28], "modeling_llama": 95, "modeling_mymodel": 95, "modeling_opt": 95, "modeling_util": [70, 95], "modelnam": 1, "modelopt": [15, 19, 54, 67, 73, 74, 86, 93], "modelopt_cuda_ext": 67, "modelpath": 0, "modelrunn": [15, 87, 93], "modelrunnercpp": [87, 93], "modelrunnermixin": 87, "modeltyp": [0, 13], "modelvari": 1, "modelweightsformat": 17, "modelweightsload": [17, 93], "modern": 87, "modif": [7, 16], "modifi": [3, 7, 65, 73, 77, 80, 81, 92, 93], "modul": [0, 1, 5, 6, 14, 15, 16, 17, 26, 29, 64, 65, 70, 80, 82, 83, 84, 86, 87, 92, 93, 95], "modular": 71, "module1": 26, "module10": 26, "module11": 26, "module12": 26, "module13": 26, "module2": 26, "module3": 26, "module4": 26, "module5": 26, "module6": 26, "module7": 26, "module8": 26, "module9": 26, "module_id": 10, "moduleid": [1, 10], "moduleidtomodel": 1, "modulelist": 95, "moduletyp": 1, "modulo": 82, "moe": [10, 17, 26, 27, 29, 50, 64, 70, 80, 82, 84, 93], "moe_4h_to_h": [10, 29], "moe_allreduce_residual_rms_norm": 82, "moe_backend": [20, 27, 70], "moe_cluster_parallel_s": 70, "moe_ep_s": 4, "moe_expert_parallel_s": [50, 70], "moe_gat": [10, 29], "moe_h_to_4h": [10, 29], "moe_load_balanc": 70, "moe_max_num_token": 70, "moe_plugin": 29, "moe_rout": [10, 29], "moe_tensor_parallel_s": [50, 70], "moe_tp_siz": 4, "moeconfig": 84, "moetopk": 93, "moment": 3, "monboardblock": 0, "monitor": [8, 29], "monitor_memori": [29, 70], "monolith": 5, "monost": 0, "month": 73, "mop": 0, "mopenipc": 1, "moptimaladapters": 0, "morchestratorconfig": 0, "morchleadercomm": 0, "more": [0, 1, 2, 3, 4, 5, 6, 7, 8, 12, 14, 15, 16, 21, 22, 23, 25, 26, 27, 28, 29, 30, 36, 40, 51, 52, 59, 65, 70, 71, 73, 74, 75, 77, 79, 80, 81, 82, 88, 89, 92, 93, 95, 97, 99], "most": [0, 1, 6, 8, 12, 16, 19, 21, 22, 23, 25, 26, 28, 42, 46, 47, 49, 50, 70, 72, 78, 80, 81, 82, 89, 92, 93], "mount": [30, 55, 56, 57], "mount_dest": [55, 56, 57], "mount_dir": [55, 56, 57], "moutdim": 1, "moutdimfirst": 1, "moutputbeamhypothes": 1, "mouttpsplitdim": 1, "move": [0, 1, 8, 19, 52, 70, 71, 82, 92, 93], "movement": [8, 16], "mownsev": 1, "mownsstream": 1, "mp4": [34, 61], "mpageblock": 1, "mpagedcontextfmha": 1, "mpagedst": 1, "mpagemanagerconfig": 1, "mpagesmutex": 1, "mpagewidth": 1, "mparallelconfig": 0, "mparticipantid": 0, "mpeftcacheconfig": 0, "mpi": [0, 1, 2, 6, 16, 18, 19, 29, 30, 55, 56, 57, 67, 70, 72, 73, 75, 82, 92, 93], "mpi4pi": [69, 75, 92, 93], "mpi_abort": 69, "mpi_barri": 19, "mpi_comm_world": [6, 69], "mpi_group_barri": 1, "mpicomm": 0, "mpicommsess": 70, "mpin": 1, "mpinneddiff": 1, "mpinnedpool": 1, "mpinnedpooldiff": 1, "mpipelineparallel": [0, 1], "mpirun": [15, 16, 69, 75, 92, 93], "mpisess": 70, "mpistat": 0, "mpointer": 1, "mpool": 1, "mport": 0, "mposteriorthreshold": 0, "mppreducescatt": 1, "mprecis": 1, "mpresencepenalti": 0, "mprocessorbatch": 0, "mprocessormap": 0, "mprompttableoffload": 0, "mpt": [25, 90, 91, 93], "mptforcausallm": 84, "mptmodel": 84, "mqa": [5, 8, 21, 24, 26, 29, 82, 93, 97], "mquantmod": 1, "mrank": [0, 1], "mrecvpollperiodm": 0, "mremotenam": 0, "mrepetitionpenalti": 0, "mreplic": 0, "mreqid": 0, "mrequeststatsmaxiter": 0, "mrnnconfig": 1, "mrope": [0, 82], "mrope_param": [83, 87], "mrope_position_delta": [82, 83, 87], "mrope_rotary_cos_sin": [82, 83], "mrope_rotary_cos_sin_s": 84, "mropeconfig": 0, "mropeparam": [83, 87], "mropepositiondelta": 0, "mroperoratysinco": 0, "mrotaryembeddingdim": 1, "mruntimedefault": 1, "mruntimestream": 1, "msamplingconfig": 1, "mscale": 82, "mscale_all_dim": 82, "mschedulerconfig": 0, "msecondaryofflineminprior": [0, 70], "msecondaryoffloadminprior": 0, "mseed": 0, "mselfidx": 0, "msg": [0, 1, 26, 70], "msinktokenlength": 0, "msizeperhead": [0, 1], "mskipcrossattnblock": 1, "msl": 1, "mslotsperpag": 1, "mspawnprocess": 0, "mspeculativedecodingconfig": 0, "mspeculativedecodingmod": 1, "mspeculativedecodingmodul": 1, "msrcdesc": 0, "mstate": [0, 1], "mstoptokenid": 0, "mstream": 1, "msyncmessag": 0, "mt5": 91, "mtag": 0, "mtaskid": 0, "mtemperatur": 0, "mtensor": 0, "mtensorparallel": [0, 1], "mtoken": 0, "mtokenizerstr": 0, "mtokenrangeretentionconfig": 0, "mtokensperblock": [0, 1], "mtopk": 0, "mtopp": 0, "mtoppdecai": 0, "mtoppmin": 0, "mtoppresetid": 0, "mtotalnumpag": 1, "mtp": [20, 70, 93], "mtp3_autoregress": 26, "mtp3_top1": 26, "mtp3_top10": 26, "mtp3_top15": 26, "mtp3_vanilla": 26, "mtpdecodingconfig": 70, "mtprank": 1, "mtransfermod": 0, "mtrimpool": 1, "mtype": [0, 1], "much": [9, 16, 27, 72, 74, 79, 89], "mul": 82, "multi": [0, 2, 3, 4, 6, 9, 10, 12, 15, 18, 19, 21, 27, 28, 29, 34, 55, 56, 57, 61, 64, 65, 69, 70, 75, 82, 84, 89, 90, 93, 97], "multi_block_mod": [5, 70, 87, 93], "multiblockmod": 0, "multidimension": 82, "multihead": [16, 21], "multimod": [0, 29, 63, 73, 87, 91, 93], "multimodalembed": 0, "multimodalmodelrunn": 87, "multinod": 76, "multinomi": 6, "multipl": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 12, 16, 17, 26, 27, 28, 29, 40, 41, 70, 71, 75, 76, 77, 79, 82, 83, 87, 88, 92, 93, 97], "multiple_profil": [29, 73, 77, 80, 93], "multipli": [5, 17, 28, 82], "multiply_and_lora": 83, "multiply_collect": 83, "multiprocessor": 16, "munsign": 1, "musecrossattent": 1, "musedynamictre": 0, "musegemmallreduceplugin": 1, "musegptattentionplugin": 1, "musegpudirectstorag": 0, "museloraplugin": 1, "musemambaconv1dplugin": 1, "musemrop": 1, "musepositionembed": 1, "museshapeinfer": 1, "musetokentypeembed": 1, "must": [0, 1, 2, 3, 4, 5, 6, 9, 10, 12, 16, 18, 29, 30, 32, 45, 70, 77, 82, 83, 85, 87, 90, 92], "mutabl": [0, 1], "mutablepageptr": 1, "mutex": [0, 1], "mutual": [6, 90], "muvm": 1, "muvmdiff": 1, "mverificationsets": 0, "mversion": 1, "mvocabs": 1, "mvocabsizepad": 1, "mweight": 0, "mwindows": 0, "mworkerexecutablepath": 0, "mworldconfig": 1, "my": [1, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 59, 66, 67, 73, 88, 94], "my_faster_on": 36, "my_model": 14, "my_profile_export": [30, 38, 39], "myattent": 95, "mybatchedlogitsprocessor": 52, "myconfig": 95, "mydecoderlay": [14, 95], "mylogitsprocessor": 52, "mymodel": [14, 95], "mymodelforcausallm": [14, 95], "n": [1, 2, 5, 10, 12, 15, 16, 27, 28, 30, 42, 45, 46, 47, 48, 49, 50, 52, 55, 56, 57, 58, 59, 69, 70, 73, 75, 79, 82, 83, 84, 89, 90, 92, 93], "n_worker": 70, "na": [73, 93], "naiv": 80, "naivepatternrewriter_replaceaddwithsub": 7, "name": [0, 1, 3, 6, 7, 10, 15, 16, 30, 31, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 56, 59, 66, 67, 69, 70, 73, 74, 75, 82, 84, 85, 86, 87, 88, 92, 93, 94, 95], "named_network_output": 92, "named_paramet": 17, "namespac": [0, 1, 69, 84], "nation": 73, "nationwid": 73, "nativ": [19, 22, 28, 93, 95], "native_quant_flow": 84, "natur": [19, 28, 34, 61, 75], "naur": [0, 3, 70], "nb": 84, "nbattentionlay": [0, 1], "nbdim": 1, "nbhead": 1, "nbkvhead": [0, 1], "nbkvheadperlay": 0, "nblayer": 1, "nbrnnlayer": 1, "nccl": [16, 26, 29, 82, 92, 93], "nccl_p2p_level": 93, "nccl_plugin": 29, "ncclplugin": 16, "ncclrecv": 82, "ncclsend": 82, "nd": [73, 82], "ndarrai": [82, 83, 87], "ndim": 82, "nearest": [28, 70, 82], "nearli": [7, 22, 28], "necess": 12, "necessari": [1, 4, 12, 26, 28, 58, 77, 82, 93, 98], "necessarili": [1, 16, 89], "need": [1, 2, 3, 5, 6, 7, 9, 12, 13, 14, 15, 16, 17, 18, 19, 20, 26, 27, 28, 30, 31, 36, 40, 45, 50, 55, 56, 57, 58, 65, 66, 67, 69, 70, 71, 73, 74, 75, 76, 77, 79, 80, 81, 82, 84, 85, 87, 88, 89, 92, 93, 95, 96, 97, 98, 99], "needed_block": 99, "needsdecoderprologu": 1, "needskvcacherewind": 1, "neg": [1, 70, 81, 82], "neglig": [9, 25, 79], "neither": [3, 82, 89], "nemo": [15, 18, 29, 71, 75, 87, 90, 91, 93], "nemo_ckpt_dir": 84, "nemo_prompt_convert": 87, "nemotron": [91, 93], "nemotron_na": 93, "nemotronforcausallm": 91, "nemotronna": [91, 93], "nemotronnasforcausallm": 91, "neox": [5, 6, 90, 91, 93], "nest": 7, "net": [9, 70, 92], "net_guard": 7, "network": [3, 4, 5, 7, 11, 16, 18, 19, 28, 29, 45, 82, 88, 89, 90, 92, 93], "neural": [4, 7, 16, 88, 93], "neva": [91, 93], "never": [7, 73, 81], "new": [0, 1, 3, 5, 6, 7, 9, 10, 12, 13, 19, 22, 23, 26, 27, 30, 31, 33, 35, 42, 46, 47, 48, 49, 50, 52, 60, 62, 64, 65, 69, 70, 71, 79, 80, 82, 87, 88, 93, 94, 96, 98], "new_decoder_architectur": [15, 84], "new_generated_id": 87, "new_input": 7, "new_out": 7, "new_shap": 82, "new_tensor": 82, "new_token": 87, "new_workflow": 93, "newactiverequestsqueuelatencym": [0, 30], "newer": [91, 93], "newest": [23, 70], "newli": [0, 70, 79], "newsiz": 1, "newtoken": 1, "newtokensstep": 1, "newtokensvec": 1, "newvalu": 0, "next": [1, 10, 12, 16, 19, 22, 27, 64, 65, 71, 76, 77, 79, 80, 81, 87, 89, 91, 93], "next_logit": 87, "next_medusa_input_id": 87, "next_medusa_logit": 87, "next_step_buff": 87, "next_step_tensor": 87, "nextdraftindic": 1, "nextdraftlen": 1, "nextdraftpath": 1, "nextdraftprob": 1, "nextdrafttoken": 1, "nextdrafttokenslen": 1, "nextflattoken": 1, "nextgenerationlength": 1, "nextn": 27, "nextpositionoffset": 1, "ngc": [66, 67, 88, 93, 94], "ngoanpv": 93, "ngram": [0, 6, 70, 84], "ngramdecodingconfig": 70, "ngramsiz": 0, "ngroup": 82, "nhead": 82, "nhere": 45, "ni": [45, 90], "nine": 88, "nj": 48, "njane": [42, 46, 47, 48, 49, 50, 52], "njason": 58, "nmh": 87, "nmt": [87, 91, 93], "nn": [82, 95], "no_quant": 70, "no_repeat_ngram_s": [6, 70, 87], "no_schedule_after_st": 99, "no_schedule_until_st": 99, "noauxtckernel": 26, "node": [0, 2, 6, 11, 18, 27, 28, 29, 55, 56, 57, 64, 69, 70, 72, 75, 76, 82, 87, 90, 92, 93], "noexcept": [0, 1], "nomin": [42, 46, 47, 48, 49, 50], "non": [0, 2, 5, 8, 13, 16, 19, 25, 26, 27, 28, 29, 52, 70, 82, 92, 93], "non_block": 52, "non_gated_vers": 82, "none": [1, 6, 7, 14, 17, 19, 29, 30, 36, 51, 52, 53, 54, 58, 59, 70, 73, 75, 79, 82, 83, 84, 85, 86, 87, 92, 93, 95, 97], "nonetyp": [70, 87], "nonzero": 82, "nor": 89, "norepeatngrams": [0, 1, 6], "norm": [17, 20, 28, 56, 72, 73, 74, 75, 82, 93, 95], "norm_before_bmm1": [83, 84], "norm_elementwise_affin": 83, "norm_ep": 83, "norm_epsilon": [15, 84], "norm_factor": 5, "norm_num_group": 83, "norm_pre_residual_weight": 82, "norm_quant_fus": 29, "norm_typ": 83, "norm_weight": 82, "normal": [0, 6, 9, 10, 13, 25, 26, 27, 28, 70, 73, 82, 89, 93], "normalize_log_prob": 70, "normalize_weight": 10, "normalized_shap": [82, 83], "normalizelogprob": [0, 1], "normedscorescba": 1, "north": [14, 16, 92], "northeastern": 88, "not_op": 82, "notabl": 25, "notat": 27, "note": [1, 2, 7, 9, 10, 11, 12, 16, 20, 23, 25, 26, 27, 28, 29, 32, 36, 51, 55, 56, 57, 59, 64, 65, 70, 73, 74, 77, 79, 81, 82, 85, 87, 89, 90, 91, 92, 94, 95, 98], "notic": [51, 58], "notifysyncmessag": 0, "notimplementederror": 19, "nougat": [90, 91, 93], "nour": 59, "now": [6, 12, 15, 17, 21, 26, 27, 71, 73, 79, 85, 88, 89, 93], "np": 82, "npy": 87, "npytorch_backend_config": 30, "nsight": 64, "nsy": 72, "ntask": [16, 30, 55, 56, 57], "null": [1, 15, 73, 88], "nullopt": [0, 1], "nullptr": [0, 1], "num": [0, 1, 20, 54, 56, 64, 70, 72, 73, 74, 75, 77, 78, 80], "num_attention_head": [15, 82, 83, 84], "num_aud_token": 87, "num_beam": [6, 87], "num_beam_group": 6, "num_block": [87, 98], "num_blocks_per_cache_level": 51, "num_bucket": [82, 83], "num_channel": [83, 84], "num_class": 83, "num_context": 97, "num_ctx_token": 97, "num_draft_token": [0, 82, 87], "num_eagle_lay": [43, 44, 70], "num_embed": 83, "num_experts_per_tok": 4, "num_gener": 97, "num_group": [82, 83], "num_head": [5, 17, 82, 87, 97], "num_hidden_lay": [15, 84, 95, 98], "num_imag": 87, "num_img_token": 87, "num_key_value_head": [15, 84, 98], "num_kv_head": [8, 82, 83, 87, 97, 98], "num_kv_heads_origin": 82, "num_kv_heads_per_cross_attn_lay": 87, "num_kv_heads_per_lay": 87, "num_lay": [82, 83, 87, 98], "num_ln_in_parallel_attn": 84, "num_local_block": 83, "num_local_expert": 4, "num_lora_module_lay": 10, "num_lora_modules_lay": 10, "num_medusa_head": [54, 70, 84, 86, 87], "num_medusa_lay": [84, 86], "num_multimodal_token": 0, "num_nextn_predict_lay": [20, 27, 70], "num_orig_po": 82, "num_po": 82, "num_postprocess_work": 30, "num_profil": 84, "num_q_head": 26, "num_request": [20, 27, 73, 74], "num_return_sequ": [87, 93], "num_sampl": 72, "num_task": 83, "num_token": [5, 26, 82, 97], "num_tokens_per_block": [82, 98], "num_tokens_per_task": 83, "num_video": 87, "numa": 11, "numacceptedtoken": 0, "numactiverequest": 0, "numattentionhead": 1, "numavailablepag": 1, "numbeamscba": 1, "number": [0, 1, 2, 3, 4, 5, 6, 8, 12, 16, 20, 24, 26, 27, 28, 29, 30, 52, 55, 56, 57, 70, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 87, 89, 90, 92, 93, 95, 97, 98], "numblockspercachelevel": 0, "numcompletedrequest": 0, "numcontextrequest": [0, 1], "numcopystream": [0, 1], "numctxsequ": 1, "numctxtoken": 0, "numdevicemodulelay": 0, "numdrafttoken": [0, 1], "numdrafttokenshost": 1, "numeaglelay": 1, "numel": 87, "numensurework": 0, "numer": [6, 11, 26, 64, 73, 88, 91], "numexpert": 1, "numgeneratedtoken": 0, "numgenrequest": 0, "numgensequ": 1, "numgentoken": 0, "numhead": 6, "numhostmodulelay": 0, "numkvattentionhead": 1, "numkvhead": 6, "numlanguag": 1, "numlay": 6, "nummissedblock": 0, "numnewactiverequest": 0, "numnewallocatedblock": 0, "numnewtokenscumsum": 93, "numnod": [0, 93], "numpag": 1, "numpausedrequest": 0, "numpi": [10, 82, 87], "numputwork": 0, "numqueuedrequest": [0, 93], "numrequestswithdrafttoken": 0, "numreturnbeam": 0, "numreturnsequ": [0, 1, 3], "numreusedblock": 0, "numscheduledrequest": 0, "numsequ": 1, "numslot": 1, "numtoken": 1, "numtotalallocatedblock": 0, "numtransformerslay": 1, "nvcc": 20, "nvcr": 93, "nvfp4": [26, 29, 59, 64, 70, 73, 93, 94], "nvidia": [15, 16, 18, 19, 20, 21, 22, 23, 25, 27, 29, 31, 33, 34, 35, 37, 38, 39, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 64, 65, 67, 71, 73, 74, 75, 80, 82, 88, 89, 91, 92, 93, 94], "nvila": [91, 93], "nvinfer1": [0, 1], "nvl": [1, 29, 93], "nvl36": 76, "nvl72": [28, 76], "nvlink": [2, 6, 11, 75, 76, 78, 93], "nvswitch": [16, 26], "nvtx": 70, "nyou": 45, "o": [0, 1, 7, 10, 19, 24, 26, 28, 55, 56, 57, 72, 92], "o_proj": 17, "oai": [34, 61], "obei": 92, "object": [0, 1, 3, 9, 14, 16, 17, 19, 36, 45, 70, 82, 83, 84, 85, 87, 88, 89, 96], "observ": [28, 51, 74], "obtain": [2, 18, 74, 82], "obviou": 28, "occas": 92, "occasion": 93, "occup": [5, 89], "occupi": [25, 28, 89], "occur": [6, 9, 98, 99], "odd": 52, "off": [9, 28, 72, 77, 79, 80, 89, 93], "offer": [16, 18, 25, 26, 71, 97], "offic": 45, "officenetsecur": 45, "offici": [5, 20, 27, 73], "offlin": [14, 23, 28, 40, 73, 74, 93], "offload": [0, 8, 13, 29, 64, 70, 93], "offset": [1, 82, 87, 90, 93], "offsetdim": 1, "ofitensor": 0, "often": [0, 3, 8, 12, 21, 25, 26, 70, 76, 77, 82], "ok": 92, "okai": 51, "old": [7, 10, 27, 92], "older": [9, 19, 65, 91], "oldest": [10, 70], "oldvalu": 0, "omit": [1, 3, 19, 82], "ompi": [67, 92], "onboard": [0, 9, 70, 89], "onboard_block": 70, "onboardblock": 0, "onc": [0, 3, 5, 6, 7, 16, 18, 65, 69, 70, 77, 82, 89], "one": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 15, 16, 17, 19, 21, 26, 27, 28, 29, 30, 31, 58, 69, 70, 73, 75, 76, 77, 80, 81, 82, 83, 85, 87, 89, 92, 93, 95, 99], "ones": [0, 10], "oneshot": [26, 82], "oneshotallreduc": 26, "oneshotar": 26, "onevis": [91, 93], "ongo": [19, 59], "onli": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 13, 14, 16, 17, 19, 20, 25, 27, 28, 29, 30, 36, 52, 59, 64, 69, 70, 73, 74, 75, 76, 77, 79, 80, 81, 82, 83, 85, 87, 89, 91, 93, 96, 99], "onlin": [18, 23, 40], "only_cross_attent": 83, "onnx": [29, 82], "onnx__gathernd": 82, "onto": 6, "oom": [20, 21, 24, 28, 89], "ootb": [28, 93], "op": [0, 1, 7, 28, 70, 82, 93], "op_and": 82, "op_or": 82, "op_xor": 82, "opaqu": 7, "opaque_st": 70, "open": [6, 21, 26, 28, 59, 71, 72, 92, 93], "openai": [30, 63, 88, 93], "openipc": 1, "openmpi": 93, "opensora": 93, "openssh": 31, "oper": [0, 1, 3, 5, 6, 7, 11, 12, 15, 16, 17, 26, 28, 29, 52, 70, 73, 76, 77, 80, 82, 88, 89, 91, 93, 96, 97, 98], "opportun": 73, "opt": [3, 15, 25, 28, 31, 82, 90, 91, 92, 93], "opt_batch_s": [70, 84], "opt_num_token": [29, 70, 84], "optforcausallm": [15, 84], "optim": [1, 2, 3, 6, 7, 8, 11, 12, 16, 18, 19, 21, 22, 23, 24, 25, 29, 46, 52, 54, 65, 69, 70, 71, 73, 74, 76, 77, 78, 82, 88, 89, 91, 92, 93, 94, 96, 97, 98], "optimaladapters": [0, 1], "option": [0, 1, 3, 6, 7, 8, 11, 12, 14, 19, 22, 27, 29, 30, 36, 52, 56, 58, 64, 67, 70, 72, 73, 74, 75, 76, 78, 79, 82, 85, 87, 89, 92, 93, 95, 97, 98], "optionalbufferptr": 1, "optionaltensorptr": 1, "optmodel": 84, "optvec": 1, "orchestr": [0, 2, 12, 92, 93], "orchestratorconfig": 0, "orchleadercomm": 0, "order": [0, 1, 2, 5, 8, 17, 21, 70, 73, 74, 77, 81, 82, 83, 89], "org": [0, 1, 4, 10, 29, 66, 67, 82, 90], "organ": [8, 71, 98], "orient": 28, "origin": [0, 5, 7, 10, 11, 27, 28, 82, 93, 95], "original_max_position_embed": [82, 83], "originaltemperatur": 1, "oserror": 93, "osl": [21, 22, 23, 24, 26, 27, 28, 73, 74, 80], "ostream": [0, 1], "other": [0, 1, 2, 3, 4, 5, 6, 9, 11, 12, 16, 17, 19, 21, 26, 27, 28, 29, 36, 49, 51, 55, 56, 57, 59, 65, 69, 70, 71, 74, 75, 76, 77, 79, 80, 81, 82, 85, 89, 92, 93, 97, 99], "other_audio_input": 87, "other_decoder_input": 87, "other_vision_input": 87, "othercach": 1, "otherwis": [0, 1, 3, 5, 6, 36, 70, 73, 82, 87, 92, 97], "our": [20, 25, 26, 27, 28, 42, 45, 46, 47, 49, 50, 73, 74, 77, 79, 80, 82, 91, 92, 93, 95], "out": [0, 1, 2, 10, 19, 21, 22, 23, 24, 26, 27, 28, 40, 55, 56, 57, 69, 72, 74, 77, 79, 80, 82, 88, 89, 93], "out_bia": 83, "out_channel": 83, "out_context_dim": 83, "out_dim": 83, "out_fatur": 15, "out_featur": [15, 16, 83], "out_hidden_s": 82, "out_of_tree_exampl": 95, "out_point": 82, "out_tp": [21, 24], "outdim": 1, "outdimfirst": 1, "outer": 82, "outlin": 72, "output": [0, 1, 2, 5, 6, 7, 9, 10, 12, 16, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 36, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 56, 58, 59, 66, 67, 70, 72, 74, 75, 76, 77, 78, 80, 81, 82, 83, 87, 88, 92, 93, 94, 96, 97, 99], "output_cum_log_prob": 87, "output_dim": 83, "output_dir": [10, 13, 14, 15, 16, 19, 29, 73, 76, 84, 86, 88, 92], "output_dtyp": [82, 83], "output_generation_logit": 87, "output_id": 87, "output_log_prob": 87, "output_multiplier_scal": 84, "output_pad": [82, 83], "output_s": 83, "output_seqlen": [21, 24], "output_sequence_length": 87, "output_timing_cach": [29, 70], "output_token": 73, "outputbuff": 1, "outputconfig": [0, 3, 36, 93], "outputidscba": 1, "outputlen": 0, "outputlogprob": 1, "outputtokenid": [0, 3], "outsid": [12, 18, 19, 97], "outsiz": 1, "outstand": 27, "outtpsplitdim": 1, "outweigh": 76, "over": [0, 1, 9, 12, 17, 20, 22, 23, 25, 26, 28, 32, 70, 72, 73, 76, 79, 80, 82, 93], "overal": [3, 5, 9, 11, 12, 20, 27, 28, 71, 76, 77, 79, 80, 81, 95], "overcom": [5, 16, 26], "overflow": 1, "overhead": [0, 3, 16, 26, 27, 28, 76, 93, 97], "overiew": 73, "overlap": [0, 2, 12, 20, 26, 27, 28, 70, 93, 99], "overload": [0, 1], "overrid": [1, 17, 19, 36, 70, 82, 87], "override_field": 84, "overshadow": 76, "oversubscrib": [69, 75], "overview": [3, 8, 20, 25, 64, 65, 72, 73, 75, 94, 96], "overwhelm": 58, "overwrit": [5, 30], "own": [0, 1, 2, 9, 12, 15, 16, 17, 18, 19, 20, 27, 36, 65, 95], "ownership": 0, "ownsev": 1, "ownsstream": 1, "p": [0, 6, 12, 18, 31, 55, 56, 57, 70, 84, 87, 93], "p2p": 82, "p50": [73, 74], "p90": [73, 74, 75], "p95": [73, 74, 75], "p99": [73, 74, 75], "p_max": 0, "p_x": 0, "pack": [0, 1, 6, 29, 64, 81, 82, 84, 89, 95], "packag": [3, 65, 66, 67, 73, 75, 92, 93], "packed_length": 84, "packedinput": 1, "packedmask": 1, "packedmaskhost": 1, "packedmaskhostcopi": 1, "packedmasksdevic": 1, "packedpositionid": 1, "pad": [0, 1, 6, 7, 10, 28, 29, 30, 64, 70, 71, 82, 83, 87, 89, 93], "pad_id": [70, 87], "pad_lda": 83, "pad_ldc": 83, "pad_token_id": 87, "padding_2d": 82, "padding_back": 82, "padding_bottom": 82, "padding_front": 82, "padding_left": 82, "padding_mod": 83, "padding_right": 82, "padding_top": 82, "padid": 0, "page": [1, 2, 6, 9, 16, 23, 29, 64, 69, 73, 75, 77, 82, 88, 89, 93, 97], "paged_context_fmha": [77, 93], "paged_kv_cach": [10, 29, 73, 87], "paged_st": [29, 87], "pagedcontextfmha": 1, "pagedkvcach": 6, "pagedst": 1, "pageid": 1, "pageidx": 1, "pagemanagerconfig": 1, "pageptr": 1, "pagewidth": 1, "pair": [0, 1, 21, 70, 77, 80, 82], "pale": 51, "paper": [2, 10, 12, 22, 27, 28, 90, 97], "par": [79, 80], "parallel": [0, 2, 3, 5, 6, 12, 15, 16, 20, 21, 23, 24, 27, 30, 40, 41, 50, 52, 64, 70, 74, 77, 78, 82, 83, 84, 89, 93, 95, 99], "parallel_attent": [15, 84], "parallelconfig": [0, 93], "param": [0, 1, 17, 46, 47, 48, 50, 51, 59, 70, 82, 83, 84, 87], "paramet": [0, 1, 3, 4, 5, 8, 9, 10, 12, 13, 15, 16, 17, 19, 20, 28, 29, 30, 55, 70, 73, 76, 77, 78, 81, 82, 83, 84, 87, 89, 93, 97], "parametr": 87, "parent": [0, 1, 17, 19], "parent_hash": 51, "parenthash": 0, "parentid": 1, "pari": [42, 46, 47, 48, 49, 50, 59], "pars": [1, 70], "parse_arg": 54, "parser": [30, 54, 63], "part": [1, 3, 4, 7, 16, 17, 19, 28, 64, 65, 69, 70, 71, 74, 79, 80, 81, 82, 87, 89], "part2": 93, "parti": 93, "partial": [0, 4, 9, 16, 26, 70, 76], "particip": [0, 59, 82, 93], "participantid": [0, 2], "particular": [0, 3, 69, 78, 79, 80, 88], "particularli": [26, 28, 65, 80, 98], "partit": [5, 10, 16, 55, 56, 57], "pass": [0, 1, 3, 5, 7, 9, 10, 12, 16, 17, 36, 52, 58, 59, 70, 72, 73, 75, 77, 79, 80, 82, 83, 84, 87, 89, 93, 94, 95, 96, 97, 99], "past": [0, 5, 27], "past_key_valu": [82, 83], "past_key_value_length": 83, "past_key_values_length": 83, "past_kv_length": 87, "past_sequence_length": 87, "patch": [83, 87], "patch_siz": [83, 84], "path": [0, 1, 3, 5, 12, 15, 17, 20, 27, 29, 30, 36, 46, 47, 48, 49, 50, 54, 55, 56, 57, 59, 65, 69, 70, 72, 73, 74, 75, 77, 82, 87, 93], "path_to_llama_from_hf": 96, "path_to_meta_llama_from_hf": 69, "path_to_trt_engin": 69, "pathlib": [54, 70], "pathlik": 84, "pathorn": 93, "pathsoffset": 1, "pattern": [4, 26, 28, 64, 70, 82, 93], "patternanalyz": 7, "patternrewrit": 7, "paus": [0, 81, 99], "paused_request": 99, "pcie": [11, 29], "pdf": [0, 4, 10], "pdl": [26, 93], "peak": [0, 20, 21, 22, 26, 74], "peft": 70, "peft_cache_config": [36, 49, 70], "peftcacheconfig": [0, 70], "peftcachemanag": [0, 93], "penal": [0, 6, 70], "penalti": 93, "penalty_alpha": 6, "pend": 99, "pending_request": 99, "per": [0, 1, 3, 5, 6, 8, 11, 12, 16, 19, 20, 21, 23, 24, 26, 27, 28, 29, 30, 55, 56, 57, 70, 73, 74, 75, 76, 77, 82, 83, 89, 90, 93], "per_channel": 90, "per_group": 90, "per_token": 90, "per_token_scal": 82, "perceiv": 22, "percent": [0, 13], "percentag": [10, 13, 73, 74, 75], "percentil": [73, 93], "perf": [0, 20, 28, 30, 63, 70, 82, 93], "perf_best_practic": 93, "perform": [0, 1, 2, 3, 5, 6, 7, 10, 16, 17, 18, 19, 21, 23, 24, 27, 28, 29, 30, 36, 65, 69, 70, 71, 73, 74, 76, 79, 81, 82, 87, 88, 91, 93, 95, 97, 98], "performantli": 21, "permut": 82, "persimmon": 93, "persist": [25, 69], "person": [31, 58], "phase": [0, 2, 7, 12, 21, 24, 26, 27, 28, 29, 64, 73, 78, 79, 80, 81, 82, 89, 93, 97, 98], "phi": [69, 82, 90, 91, 93], "phi3config": 84, "phi3forcausallm": 84, "phi3model": 84, "phiconfig": 84, "phiforcausallm": 84, "phimodel": 84, "physic": [82, 89], "picasso": 59, "pick": 79, "pickl": 93, "piec": 79, "piecewis": 70, "pin": [0, 1, 9], "ping": 93, "pinnedmemusag": 0, "pinnedpool": 1, "pip": [20, 30, 65, 66, 67, 88, 93], "pip3": [66, 67], "pipelin": [0, 1, 3, 6, 16, 21, 24, 29, 30, 50, 70, 73, 74, 78, 89, 93, 99], "pipeline_parallel_s": [50, 70, 76, 77], "pipelineparallel": [0, 1, 6], "pipelineparallelismrank": 1, "pitfal": [9, 19], "pixart": 83, "pixartalphatextproject": 83, "pixel_valu": 84, "pl": [67, 73], "place": [1, 29, 51, 67, 82, 93, 95], "placement": 26, "plai": 79, "plan": [3, 5, 26, 65], "planner": 93, "platform": [31, 32, 42, 46, 47, 49, 50, 65, 71, 73, 93, 94], "pleas": [2, 5, 7, 11, 12, 14, 21, 23, 24, 25, 26, 28, 32, 36, 45, 52, 65, 67, 73, 74, 76, 78, 82, 92, 93, 94, 99], "plu": [11, 87], "plugin": [5, 6, 7, 13, 15, 64, 65, 70, 79, 82, 84, 88, 89, 90, 92, 93], "plugin_config": [70, 77, 80, 82, 84], "plugin_namespac": 7, "plugin_typ": 7, "plugin_v2": 7, "plugin_v2_gemm_0": 92, "pluginconfig": [70, 85], "pluginconfigmeta": 85, "pluginfield": 93, "pluginv2build": 92, "pm": [20, 26, 73], "pmi": 92, "pmi2_init": 92, "pmix": [16, 30, 55, 56, 57, 92], "png": [34, 39, 61], "po": 83, "point": [1, 5, 16, 18, 22, 25, 40, 45, 50, 66, 67, 69, 70, 74, 76, 81, 82, 88, 90, 92, 93], "pointer": [0, 1, 6, 17, 82, 87, 93], "pointerelementtyp": 1, "polar": 91, "polici": [0, 1, 2, 70, 73, 75, 89], "poll": [0, 30], "polyhedr": 16, "pong": 93, "pool": [0, 1, 5, 28, 64, 70, 82, 87, 98, 99], "pooled_project": [83, 84], "pooled_projection_dim": 83, "pooledpin": 0, "poor": 2, "popd": 92, "popfirstgentoken": 0, "popul": [1, 5, 16, 59, 82], "popular": [5, 15, 19, 25, 27, 32, 69], "port": [0, 30, 32, 37], "portfolio": 23, "portion": [4, 76, 82, 89], "pos_emb_typ": 82, "pos_embd_param": 97, "pos_embed_max_s": 83, "pos_embed_typ": 83, "pose": 80, "posit": [0, 1, 12, 26, 27, 70, 73, 82, 83, 87, 93, 97], "position_embed": [82, 83], "position_embedding_typ": [5, 15, 82, 83, 84], "position_encoding_2d": 84, "position_id": [84, 87, 92, 95, 97], "positionalembeddingparam": 97, "positionembeddingtyp": [5, 82, 83, 84], "positionid": [0, 1], "positionidsbas": 1, "positionidsdevic": 1, "positionidshost": 1, "positionidshostcopi": 1, "positionoffset": 1, "positionoffsetsdevic": 1, "positionoffsetshost": 1, "positionoffsetshostcopi": 1, "posix": 0, "posix_debug_fallback": 0, "possibl": [2, 3, 5, 6, 9, 12, 16, 20, 27, 28, 29, 36, 65, 71, 72, 73, 74, 77, 79, 81, 82, 89, 92, 93, 96], "possibli": [1, 8, 82], "post": [0, 15, 22, 25, 26, 27, 28, 59, 71, 72, 82, 88, 93], "post_act_fn": 83, "post_attention_layernorm": [17, 95], "post_input_id": 87, "post_layernorm": [14, 15, 17, 82, 92], "post_pad": 82, "post_prompt": 87, "post_strid": 82, "posterior_threshold": [43, 44, 70], "posterioralpha": 1, "posterioralphahost": 1, "posteriorthreshold": [0, 1], "posteriorthresholdhost": 1, "postprocess": [30, 83], "postprocessor": [0, 70], "postprocparam": 70, "potenti": [0, 1, 8, 12, 28, 29, 72, 73, 77, 95], "pow": 82, "power": [9, 16, 23, 25, 26, 28, 71, 79, 93], "pp": [0, 2, 6, 10, 21, 24, 30, 73, 75, 82, 93], "pp2": 73, "pp_communicate_final_output_id": 87, "pp_communicate_new_token": 87, "pp_reduce_scatt": [29, 80], "pp_size": [15, 16, 30, 37, 73, 74, 76, 86, 93], "ppreducescatt": 1, "pr": 26, "practic": [5, 8, 16, 22, 23, 26, 28, 88, 89, 93], "pre": [0, 1, 3, 5, 15, 18, 65, 67, 70, 71, 73, 82, 88, 89, 93, 97], "pre_input_id": 87, "pre_layernorm": 82, "pre_onli": 83, "pre_pad": 82, "pre_prompt": 87, "pre_quant_scal": [15, 70], "pre_strid": 82, "prebuilt": 65, "preced": [16, 82], "precis": [1, 6, 17, 21, 25, 29, 64, 73, 77, 80, 85, 88, 89, 91, 93], "precompute_relative_attention_bia": 84, "precomputed_relative_attent": 83, "predefin": [12, 95, 97], "predict": [1, 5, 12, 26, 27, 93], "predicteddraftlogit": 1, "predictor": 12, "predictsdrafttoken": 1, "prefer": [25, 65], "prefer_managed_weight": 83, "prefer_plugin": 82, "prefetch": 26, "prefil": [0, 28, 70, 78], "prefix": [3, 12, 15, 27, 51, 69, 75, 82, 85, 92], "preliminari": [21, 23, 24], "preload": 17, "premis": 27, "prepar": [0, 1, 2, 26, 27, 51, 56, 64, 72, 79, 82, 84, 90, 93, 97], "prepare_dataset": [20, 56, 72, 73, 74, 75], "prepare_input": [84, 89], "prepare_position_ids_for_cogvlm": 87, "prepare_recurrent_input": 84, "prepare_resourc": [96, 98], "prepareforward": 1, "prepend": 92, "preprocess": [17, 87, 90], "preprocess_weights_hook": 84, "preprocessor": 73, "prequant_scaling_factor": 15, "prerequisit": [64, 66, 67], "presenc": [6, 16, 51], "presence_penalti": [70, 87, 93], "presencepenalti": [0, 1, 6], "present": [0, 70, 73, 79, 80, 90, 93], "preserv": 77, "presid": [40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 59, 66, 67, 75, 81, 88, 94], "pretrain": 18, "pretrained_config": 95, "pretrained_model_name_or_path": 84, "pretrainedconfig": [14, 19, 70, 84, 85, 95], "pretrainedmodel": [19, 84, 89], "pretrainedtokenizerbas": 70, "prevdrafttokenslen": 1, "prevent": [26, 28, 64, 69], "preview": 93, "previou": [1, 3, 4, 12, 19, 20, 22, 27, 73, 75, 76, 77, 79, 80, 81, 93], "previous": [1, 21, 77, 79, 81, 93], "prevscor": 1, "prewritten": 88, "price": 73, "primari": [0, 1, 8, 25, 89, 99], "primarili": 97, "primit": [16, 28, 71, 88], "print": [1, 5, 30, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 60, 61, 62, 66, 67, 70, 73, 74, 75, 81, 88, 89, 92, 94], "print_iter_log": [20, 56, 70], "prior": [3, 29, 65, 67], "priorit": [25, 79, 81], "prioriti": [0, 1, 8, 9, 17, 70], "prioritytyp": 0, "priorityupd": 0, "privat": [0, 1, 6, 70], "privileg": 7, "prm": 91, "pro": 26, "prob": 82, "probabilist": 83, "probabl": [0, 1, 6, 9, 12, 26, 27, 70, 82, 87, 93], "probil": 1, "problem": [5, 20, 28, 92], "proc": 17, "proccessed_weight": 17, "proccessed_zero": 17, "procedur": 20, "proceed": 16, "process": [0, 1, 2, 3, 5, 6, 8, 11, 12, 15, 16, 19, 20, 26, 27, 28, 29, 40, 45, 50, 52, 55, 56, 57, 66, 67, 69, 70, 71, 72, 73, 74, 75, 76, 79, 80, 81, 82, 87, 88, 92, 93, 95, 96, 97, 99], "process_input": 87, "process_logits_including_draft": 87, "processor": [0, 5, 40, 41, 53, 70, 84, 87, 93], "processorbatch": 0, "processormap": 0, "prod": 82, "produc": [0, 1, 3, 7, 16, 36, 73, 75, 77, 79, 80, 82, 93], "product": [4, 5, 12, 16, 23, 71, 79, 80, 81, 82, 88, 97], "profil": [2, 29, 30, 38, 39, 64, 77, 79, 82, 87, 89, 92, 93], "profiling_verbos": [29, 70], "profit": [12, 73], "program": [2, 19, 40, 42, 46, 47, 49, 50, 52, 66, 67, 69, 81, 88, 92], "progress": [1, 26, 70, 73, 82], "proj": [15, 17, 92], "project": [5, 10, 28, 59, 65, 82, 83, 95, 98], "projector_hidden_act": 84, "prologu": [55, 56, 57], "promin": 12, "promis": [12, 19, 27], "prompt": [0, 3, 6, 9, 14, 20, 29, 30, 35, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 58, 59, 62, 64, 66, 67, 70, 73, 75, 79, 80, 81, 83, 87, 88, 93, 94, 97], "prompt_adapter_request": [70, 93], "prompt_embedding_t": [83, 84, 87], "prompt_embedding_table_s": 84, "prompt_id": 52, "prompt_len": 97, "prompt_logprob": 70, "prompt_lookup": [12, 93], "prompt_lookup_num_token": [6, 70], "prompt_tabl": 87, "prompt_task": [84, 87], "prompt_token": 88, "prompt_token_id": [36, 53, 70], "prompt_vocab_s": [84, 87], "promptadapterrequest": 70, "promptinput": [70, 93], "promptlen": 0, "prompttableoffload": 0, "prompttuningconfig": 0, "prompttuningembed": 83, "prompttuningen": 1, "pronounc": 12, "proof": 98, "propag": [9, 93], "proper": [2, 73], "properli": [17, 79, 81], "properti": [3, 45, 70, 82, 84, 85, 87], "proport": 5, "propos": [0, 26], "protect": [1, 40, 50, 66, 67, 69, 88], "protocol": [0, 30, 45], "proud": 26, "prove": [12, 28], "provid": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 18, 19, 20, 21, 22, 25, 26, 28, 29, 30, 31, 36, 45, 54, 59, 65, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 82, 87, 89, 91, 92, 93, 95, 96, 97], "proxy_dispatch_result_thread": 73, "prune": [7, 12, 82], "pseudo": [5, 82, 90], "pth": [17, 93], "ptq": [25, 77, 93], "ptr": 1, "ptr_idx": 17, "ptrdiff_t": 1, "ptuning_setup": 87, "ptuning_setup_fuyu": 87, "ptuning_setup_llava_next": 87, "ptuning_setup_phi3": 87, "ptuning_setup_pixtr": 87, "ptuningconfig": 0, "public": [0, 1, 25, 32, 54, 59], "publish": [20, 21, 24, 73, 74, 93], "pull": [18, 20, 65, 88, 93], "puneeshkhanna": 93, "purchas": 73, "pure": 87, "purpos": [5, 8, 28, 65, 75, 77, 79, 80], "pursu": [42, 46, 47, 49, 50, 52], "push": [28, 31, 53], "pushd": 92, "put": [1, 15, 26, 55, 56, 57, 69, 71, 79], "pwd": [20, 65], "py": [3, 4, 5, 7, 10, 12, 13, 14, 15, 16, 17, 19, 20, 26, 27, 28, 52, 55, 56, 65, 67, 69, 72, 73, 74, 75, 76, 77, 82, 85, 87, 88, 92, 93, 95, 96, 98, 99], "py3": 93, "py_executor_cr": 99, "pybind": 93, "pybind11_object": 70, "pybindmirror": 70, "pydant": [70, 93], "pydantic_cor": 70, "pyexecutor": [93, 98, 99], "pynvml": 93, "pypi": [65, 93], "python": [1, 5, 6, 7, 10, 12, 14, 16, 18, 19, 20, 27, 28, 30, 36, 47, 48, 64, 66, 67, 69, 72, 73, 74, 75, 76, 88, 90, 93, 95, 96, 98, 99], "python3": [10, 13, 15, 20, 55, 56, 65, 67, 72, 73, 88, 92], "python_bind": 20, "python_e2": 87, "python_plugin": 93, "pythonpath": [20, 56, 57], "pytorch": [7, 12, 15, 18, 20, 27, 28, 30, 37, 51, 55, 56, 57, 64, 65, 66, 67, 70, 74, 82, 93, 96, 97, 98, 99], "pytorch_backend_config": 30, "pytorch_eagle_weights_path": 70, "pytorch_extra_arg": 56, "pytorch_model": 92, "pytorch_model_engin": 96, "pytorch_model_registri": 98, "pytorchconfig": [70, 97], "pytorchmodelengin": [96, 98], "pzzzzz5142": 93, "q": [2, 5, 6, 10, 21, 26, 28, 64, 73, 82, 92, 95, 97], "q_b_proj": 82, "q_dim": 82, "q_lora_rank": [82, 83], "q_proj": [17, 95], "q_scale": [5, 82, 83, 84], "qa": 12, "qformat": [73, 86], "qgmma": 93, "qingquansong": 93, "qk_layernorm": [83, 84], "qk_nope_head_dim": [82, 83], "qk_norm": 83, "qk_rope_head_dim": [82, 83], "qkv": [7, 10, 15, 17, 64, 82, 92, 93, 97], "qkv_bia": [82, 93], "qkv_dim": 82, "qkv_proj": 95, "qo_indptr": 97, "qpi": 11, "qserv": 93, "quadrat": [5, 89], "qualiti": [27, 77, 80], "qualnam": [70, 82, 84, 86], "quant": [19, 70, 73, 82, 93, 94], "quant_algo": [15, 17, 19, 36, 59, 70, 73, 77, 84], "quant_and_calib_config": 59, "quant_config": [19, 36, 59, 70, 77, 84, 97], "quant_medusa_head": 86, "quant_mod": [19, 70, 83, 84, 87], "quantalgo": [36, 59, 70, 77, 84, 86], "quantconfig": [19, 36, 59, 70, 77, 84, 93, 97], "quanticonfig": 19, "quantiz": [5, 6, 11, 16, 17, 20, 21, 22, 26, 28, 29, 40, 41, 46, 54, 64, 67, 68, 69, 70, 71, 74, 75, 78, 82, 83, 84, 87, 88, 91, 93, 95, 97], "quantizaton": 73, "quantize_and_export": 86, "quantize_kwarg": 84, "quantize_lm_head": [86, 93], "quantized_valu": 5, "quantizedkernel": 16, "quantizetensorplugin": 16, "quantmod": [1, 5, 6, 64, 70, 82, 83, 84, 86, 87], "quantmodewrapp": [70, 82], "queri": [3, 6, 8, 12, 16, 21, 28, 30, 64, 73, 82, 89, 97, 98], "query_dim": 83, "query_key_valu": 17, "query_length": 83, "query_pre_attn_scalar": 84, "question": [58, 73, 89, 92], "queu": [0, 74, 79], "queue": [0, 70, 71, 96], "quick": [5, 64, 71, 73, 75, 97], "quick_gelu": 82, "quicker": 76, "quickli": [19, 88], "quickstart": [69, 75], "quickstart_advanc": [27, 55], "quit": [7, 69], "qweight": 17, "qwen": [17, 30, 39, 69, 73, 82, 90, 91, 93], "qwen1": [91, 93], "qwen2": [10, 30, 34, 39, 61, 73, 91, 93], "qwen2_5_vlforconditionalgener": 91, "qwen2audio": 93, "qwen2forcausallm": 91, "qwen2forprocessrewardmodel": 91, "qwen2forrewardmodel": 91, "qwen2forsequenceclassif": 93, "qwen2vl": 93, "qwen2vlforconditionalgener": 91, "qwenforcausallm": 17, "qwenforcausallmgenerationsess": 87, "qwenvl": 93, "qwq": 91, "qychen": 10, "qzero": 17, "r": [1, 10, 30, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 66, 67, 75, 81, 82, 88, 92, 93, 94], "r1": [30, 63, 74, 93], "r1_in_tensorrt": [26, 93], "race": 93, "radix": 98, "rai": 1, "rais": [19, 70, 75, 92, 93], "rand": 82, "rand_data": 82, "rand_data_sampl": 84, "rand_data_valid": 84, "random": [0, 6, 30, 38, 39, 70, 74, 82, 93], "random_se": [70, 84, 87], "randomdatasampl": 1, "randomdatavalid": 1, "randomli": 74, "randomse": [1, 6, 93], "randomseedtyp": 0, "rang": [0, 6, 9, 12, 72, 80, 82, 84, 89, 90, 91, 92, 95], "rank": [0, 1, 2, 3, 4, 6, 10, 19, 20, 28, 29, 69, 73, 82, 84, 87, 89, 92, 93], "rank0": 15, "rank1": 15, "rapid": [12, 74, 88], "rate": [0, 20, 26, 27, 28, 30, 38, 39, 73, 74, 75, 93], "rather": [5, 7, 12, 28, 67, 71], "ratio": 28, "rational": 28, "raw": 30, "raw_audio": 87, "raw_imag": 87, "rdma": 2, "re": [20, 25, 70, 71, 93, 97], "reach": [0, 5, 15, 69, 73, 77, 81], "read": [0, 2, 3, 5, 12, 14, 16, 17, 20, 26, 27, 29, 58, 70, 73, 93], "read_config_from_the_custom_training_checkpoint": 19, "readabl": 73, "reader": 82, "readi": [0, 88], "readm": [2, 12, 30, 69, 75, 93], "real": [7, 20, 26, 65, 75, 77, 79, 80, 82, 92], "realiti": 79, "realiz": [9, 12], "rearrang": 82, "reason": [0, 5, 6, 16, 19, 26, 27, 30, 63, 70, 73, 76, 79, 80, 82, 92], "reasoning_pars": [30, 37], "rebuild": [80, 82, 92], "receiv": [0, 1, 2, 3, 4, 11, 12, 77, 82, 93], "recent": [1, 4, 5, 22, 26], "recip": [26, 28, 30, 70, 90], "reclaim": 0, "recogn": [12, 26, 73, 95], "recommend": [2, 5, 6, 12, 14, 17, 18, 20, 22, 25, 28, 30, 52, 65, 70, 73, 78, 79, 81, 92, 93, 95, 97], "recompute_scale_factor": 82, "reconfigur": [3, 67], "reconstruct": [5, 82], "record": [1, 7, 20, 26, 27, 70], "recored": 0, "recreat": 18, "recurr": 12, "recurrentgemma": [90, 91, 93], "recurrentgemmaforcausallm": 84, "recurs": [20, 65, 69], "recv": [0, 16, 82], "recvconnect": 0, "recvpollperiodm": 0, "recycl": [5, 98], "redesign": 93, "redirect": [7, 70], "redraft": [64, 82, 87, 93], "redrafter_draft_len_per_beam": 87, "redrafter_inverted_temperatur": 84, "redrafter_num_beam": 87, "redrafterforcausallm": 84, "reduc": [2, 3, 4, 5, 9, 11, 12, 16, 20, 21, 24, 26, 27, 28, 29, 65, 69, 71, 72, 73, 74, 75, 76, 79, 81, 82, 89, 92, 93, 97], "reduce_fus": [29, 73, 77, 80], "reduce_scatt": 82, "reduceoper": 82, "reducescatt": [29, 80, 93], "reduct": [11, 12, 26, 81, 82], "redund": [12, 26], "refactor": [19, 27, 93], "refer": [0, 1, 2, 3, 5, 6, 7, 8, 10, 12, 16, 18, 19, 20, 30, 32, 33, 34, 35, 36, 37, 38, 39, 40, 52, 60, 61, 62, 65, 69, 71, 73, 74, 75, 76, 77, 78, 80, 82, 88, 91, 93, 95, 97], "referenc": 77, "reference_wrapp": [0, 3], "refin": 93, "refit": [16, 29, 93], "refit_engin": 16, "reflect": 79, "refresh": 73, "regard": 82, "regardless": 92, "regex": [3, 70], "region": 72, "regist": [31, 64, 92, 93, 95], "register_auto_model": 95, "register_network_output": 92, "registerdesc": 0, "registermemori": 0, "regress": [5, 6, 16], "regular": [0, 3, 5, 26, 70, 82], "reinforc": 78, "reject": [0, 27], "rel": [9, 21, 79, 81, 82, 93], "rel_attn_t": 83, "relat": [2, 4, 8, 17, 64, 71, 72, 82, 85, 89, 92, 93, 94, 95, 98], "relationship": 89, "relative_attent": [82, 83], "relative_attention_bia": 82, "relax": 5, "relaxed_delta": [26, 27, 70], "relaxed_topk": [26, 27, 70], "releas": [1, 5, 6, 8, 19, 21, 24, 25, 64, 71, 82, 84, 89, 90, 91], "release_build": 65, "release_run": [65, 88], "releasepag": 1, "releasest": 0, "relev": [6, 65, 98], "reli": [2, 5, 7, 19, 69, 72, 90], "reload": 3, "relu": [15, 16, 82, 92], "remain": [0, 7, 9, 12, 13, 26, 65, 74, 75, 77, 79, 80, 82, 89, 93], "remaind": 77, "remark": [26, 27], "remind": [5, 97], "remot": 70, "remotenam": 0, "remov": [0, 1, 5, 6, 7, 8, 16, 17, 20, 27, 29, 30, 54, 65, 70, 71, 77, 82, 89, 93, 95], "remove_const_t": 1, "remove_cv_t": 0, "remove_duplicated_kv_head": 84, "remove_input_pad": [5, 10, 29, 82, 83, 87], "remove_pointer_t": 1, "remove_reference_t": 1, "remove_sequ": 98, "renam": 93, "reorder": [82, 83], "reorder_kv_cache_for_beam_search": 87, "rep": 72, "repeat": [0, 5, 27, 28, 70, 82], "repeat_interleav": 82, "repeatedli": 12, "repetit": [0, 6, 70, 82], "repetition_penalti": [6, 70, 87, 93], "repetitionpenalti": [0, 1, 6], "replac": [1, 4, 7, 16, 17, 19, 20, 28, 73, 75, 77, 81, 82, 89, 95], "replace_add_with_sub": 7, "replace_all_uses_with": [7, 82], "replace_input_with": 7, "replace_output_uses_with": 7, "replace_outputs_uses_with": 7, "replic": [0, 3, 26, 82], "replit": [90, 91, 93], "repo": [19, 69, 71, 75, 92], "repo_id": 58, "report": [8, 27, 28, 72, 73, 74, 89, 93], "reportpluginerror": 92, "repositori": [12, 18, 20, 31, 69, 88], "repres": [0, 1, 2, 8, 12, 20, 21, 25, 26, 45, 58, 70, 73, 79, 82, 87, 99], "represent": [7, 16], "reproduc": [64, 73, 93], "req": [20, 73, 74, 75, 77, 79, 80], "req_id": 52, "req_logit": 52, "req_stat": 99, "req_token_id": 52, "reqbeamwidth": 1, "reqid": 0, "reqpromptlength": 1, "request": [0, 2, 5, 6, 9, 10, 16, 20, 22, 24, 27, 28, 29, 30, 38, 39, 52, 56, 70, 71, 72, 73, 74, 75, 77, 79, 80, 81, 82, 88, 89, 93, 96, 97, 98, 99], "request_id": [36, 53, 70, 97], "request_stats_max_iter": 70, "request_timeout": 30, "request_typ": 70, "request_type_context_and_gener": [0, 2], "request_type_context_onli": [0, 2], "request_type_generation_onli": [0, 2], "requesterror": 70, "requestid": [0, 2, 3], "requestidtyp": 0, "requestlist": 99, "requestoutput": [36, 53, 70, 93], "requestperfmetr": 0, "requestschedul": 99, "requeststag": 0, "requeststat": 0, "requeststatsmaxiter": 0, "requeststatsperit": 0, "requeststatsperiter": 0, "requeststatsvec": 0, "requesttoken": 3, "requesttyp": [0, 1, 2, 70], "requesttypesdevic": 1, "requestvector": 1, "requir": [0, 2, 5, 6, 9, 10, 12, 16, 17, 19, 20, 21, 25, 26, 28, 29, 30, 45, 58, 65, 66, 67, 70, 73, 74, 75, 76, 77, 80, 82, 83, 88, 89, 91, 92, 93, 98], "require_ln_f": 84, "requiresattentionmask": 1, "rerun": 80, "rescale_output_factor": 83, "research": [5, 27, 32, 42, 46, 47, 49, 50, 90], "resembl": 51, "reserv": [0, 1, 30, 70, 81, 87, 89, 99], "reserved_block": 99, "reset": [0, 1, 6, 70, 73, 87], "resetspeculativedecodingmodul": 1, "reshap": [1, 82], "resid": [10, 59], "residu": [82, 92], "residual_connect": 83, "residual_mlp": 84, "residual_multipli": 84, "residual_rms_norm": 82, "residual_rms_norm_out_quant_fp8": 82, "residual_rms_norm_out_quant_nvfp4": 82, "residual_rms_norm_quant_fp8": 82, "residual_rms_norm_quant_nvfp4": 82, "residual_rms_prepost_norm": 82, "residualadd": [29, 80, 93], "resiz": 1, "resolv": [34, 61, 92], "resourc": [0, 2, 5, 19, 26, 28, 96, 98, 99], "respect": [4, 36, 81, 82, 87, 89, 90, 95, 99], "respons": [0, 2, 8, 30, 36, 60, 61, 62, 70, 73, 82, 96], "responsewithid": 0, "rest": [1, 5, 77], "restart": 0, "restrict": [0, 2, 3, 6, 65, 70, 82], "result": [0, 1, 4, 5, 11, 12, 16, 21, 22, 23, 25, 27, 28, 29, 36, 64, 65, 70, 73, 76, 77, 78, 79, 80, 82, 83, 93, 95, 97, 99], "retail": 73, "retain": [21, 23], "retent": [0, 70], "retentionprior": 0, "retentionpriorityanddur": 0, "rethink": 12, "retriev": [1, 17, 70, 74, 82], "return": [0, 1, 3, 7, 10, 12, 14, 16, 17, 19, 36, 70, 73, 79, 82, 83, 84, 87, 89, 92, 93, 98, 99], "return_all_generated_token": 87, "return_context_logit": 70, "return_dict": 87, "return_encoder_output": [70, 87], "return_generation_logit": 70, "return_perf_metr": 70, "returnallgeneratedtoken": [0, 3], "returncontextlogit": 0, "returnencoderoutput": 0, "returngenerationlogit": 0, "returnlogprob": 0, "returnperfmetr": 0, "reus": [0, 2, 3, 8, 27, 29, 64, 68, 70, 82, 87, 89, 93, 95, 98], "reusabl": [8, 9], "reusedblock": 0, "reusedblocksperrequest": 0, "reveal": [26, 28], "revers": 82, "revert": 82, "review": 73, "revis": 70, "revolution": 71, "rewind": [27, 93], "rewrit": [64, 82, 93, 95], "rewritepatternmanag": 7, "rewrt": 92, "rf": 92, "rg_lru": 82, "rgc": 73, "rh": [0, 1], "rich": 15, "right": [71, 77, 82, 92], "rigor": [51, 73], "risk": [2, 16, 77, 81], "rm": [65, 82, 91, 92, 95], "rms_norm": [26, 82, 95], "rmsnorm": [10, 26, 82, 83, 84, 93, 95], "rnn": [29, 93], "rnn_conv_dim_s": 87, "rnn_head_siz": 87, "rnn_hidden_s": 87, "rnn_state": 84, "rnnconfig": 1, "rnnconvdims": 1, "rnnheadsiz": 1, "rnnhiddens": 1, "ro": 20, "roberta": [91, 93], "robertaforquestionansw": 84, "robertaforsequenceclassif": 84, "robertamodel": 84, "robin": 2, "robust": [26, 93], "rock": 82, "role": [16, 30, 33, 34, 45, 60, 61, 79, 88], "roll": 64, "rooflin": 28, "root": [15, 20, 31, 65, 67, 69, 70, 75, 82, 88], "root_lay": 7, "rope": [26, 28, 82, 87, 93, 97], "rope_gpt_neox": [5, 82, 84], "rope_gptj": [5, 82], "rope_local_base_freq": 84, "rope_scaling_config": 82, "rope_scaling_long_factor": 83, "rope_scaling_long_mscal": 83, "rope_scaling_short_factor": 83, "rope_scaling_short_mscal": 83, "ropeembeddingutil": 82, "rotari": [0, 26, 82, 87, 95, 97], "rotary_bas": 84, "rotary_cos_sin": 82, "rotary_dim": 84, "rotary_embed": 95, "rotary_embedding_bas": [82, 83], "rotary_embedding_base_loc": 83, "rotary_embedding_beta_fast": 83, "rotary_embedding_beta_slow": 83, "rotary_embedding_dim": [5, 82, 84], "rotary_embedding_long_m_scal": 82, "rotary_embedding_max_posit": 82, "rotary_embedding_mscal": 83, "rotary_embedding_mscale_all_dim": 83, "rotary_embedding_origin_max_posit": 83, "rotary_embedding_original_max_posit": 82, "rotary_embedding_percentag": 83, "rotary_embedding_sc": 83, "rotary_embedding_scal": 82, "rotary_embedding_scale_typ": 82, "rotary_embedding_short_m_scal": 82, "rotary_inv_freq": [82, 83], "rotary_inv_freq_loc": 83, "rotary_pct": 84, "rotary_sc": [83, 84], "rotaryembed": 95, "rotaryembeddingdim": [0, 1], "rotaryscalingtyp": 82, "rotate_every_two": 82, "rotate_half": 82, "round": [2, 70, 82], "rout": [2, 28], "router": [4, 10, 28, 93], "router_gemm": 26, "routin": 7, "routingkernel": 26, "row": [10, 79, 82, 90, 93], "rowlinear": [10, 83], "rowwis": 70, "rr": 93, "rslora": 93, "rst": 3, "rtx": 93, "rubric": 82, "rule": [5, 76, 92], "run": [0, 1, 2, 3, 5, 6, 9, 11, 12, 14, 15, 16, 21, 25, 26, 28, 29, 30, 31, 32, 47, 48, 52, 55, 56, 57, 64, 65, 66, 67, 69, 70, 71, 76, 77, 79, 80, 81, 82, 84, 87, 89, 90, 92, 93, 95, 96, 97, 98], "run_dtm_pld": 12, "run_medusa_decod": 54, "runner": [0, 15, 87], "runningleon": 93, "runpod": 31, "runtim": [0, 3, 5, 12, 13, 18, 26, 27, 29, 30, 49, 52, 58, 64, 70, 71, 72, 73, 75, 78, 79, 82, 83, 84, 88, 92, 93, 95, 97, 99], "runtime_config": [36, 49], "runtime_default": 84, "runtime_error": 1, "runtime_rank": 87, "runtimedefault": [0, 84], "runtimedefaultsin": 84, "runtimeerror": [69, 70, 92], "runtimetensor": 87, "s0": 5, "s1": 5, "s2": 5, "sacrif": 26, "sad": 87, "saeyoonoh": 93, "safe": [1, 7, 28, 80], "safer": 82, "safetensor": [15, 17, 92, 93], "sage_attn": 82, "sage_attn_k_block_s": 82, "sage_attn_k_quant_s": 82, "sage_attn_q_block_s": 82, "sage_attn_q_quant_s": 82, "sage_attn_v_block_s": 82, "sage_attn_v_quant_s": 82, "sageattent": 82, "sai": [72, 75, 79], "said": 77, "sake": 79, "sale": 73, "same": [0, 1, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 16, 19, 22, 27, 28, 29, 52, 55, 56, 57, 65, 69, 70, 73, 74, 77, 80, 81, 82, 83, 85, 87, 89, 93], "sampl": [0, 1, 3, 5, 16, 18, 20, 26, 27, 43, 44, 46, 47, 48, 49, 50, 51, 52, 54, 58, 59, 64, 68, 70, 72, 73, 74, 82, 83, 87, 93], "sample_proj_bia": 83, "sample_weight_strip": 93, "samplemod": 82, "sampler": 70, "sampling_config": 87, "sampling_param": [36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 59, 66, 67, 70, 75, 81, 88, 93, 94], "samplingconfig": [0, 3, 6, 36, 87, 93], "samplingparam": [36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 59, 66, 67, 70, 75, 81, 88, 93, 94], "saniti": [66, 67, 76, 77, 80], "santacod": [69, 90, 91], "satfinit": 90, "satisfi": [6, 17, 93], "save": [5, 9, 12, 19, 20, 27, 28, 29, 31, 46, 49, 69, 70, 72, 73, 77, 80, 81, 89, 93], "save_checkpoint": [19, 84], "save_config": [19, 84], "saw": [77, 88], "sbatch": [16, 55, 56, 57], "sbsa": [93, 94], "scaffold": [93, 95], "scalar": [6, 11, 82], "scalartyp": 93, "scale": [0, 6, 10, 17, 28, 29, 70, 77, 82, 83, 90, 93], "scale_d0": 82, "scale_d1": 82, "scale_factor": 82, "scale_output": 82, "scale_qk": 83, "scale_typ": 82, "scalia": [42, 46, 47, 49, 50], "scaling_factor": 82, "scaling_long_factor": 82, "scaling_short_factor": 82, "scalingvecpoint": 1, "scanreducetempstorag": 1, "scanreducetempstoragebyt": 1, "scantempstorag": 1, "scantempstoragebyt": 1, "scatter": [7, 82], "scatter_nd": 82, "scenario": [2, 5, 11, 12, 15, 20, 23, 25, 26, 28, 29, 32, 73, 74, 75, 77, 79, 80, 93], "scfg": 87, "schedul": [0, 2, 3, 9, 10, 20, 27, 28, 29, 30, 51, 70, 73, 75, 80, 89, 93, 94], "schedule_request": 99, "scheduled_request": 99, "scheduler_config": [70, 81], "schedulerconfig": [0, 70, 81, 93], "schedulerpolici": 93, "schema": [0, 3, 45, 70, 73], "scheme": 0, "scicod": 26, "scienc": [42, 46, 47, 49, 50, 52], "scope": [18, 27, 93], "score": [6, 28], "scout": 91, "scratch": [73, 75, 76, 80], "script": [10, 14, 16, 19, 20, 31, 55, 56, 57, 65, 69, 72, 73, 74, 75, 85, 90, 92, 93, 94, 95], "sd3": 83, "sd35adalayernormzerox": 83, "sd3patchemb": 83, "sd3transformer2dmodel": 84, "sd3transformer2dmodelconfig": 84, "sdxl": 93, "seamless": 93, "search": [0, 1, 3, 6, 12, 18, 24, 29, 30, 36, 49, 64, 70, 77, 79, 82, 93, 96], "seashor": [34, 61], "seat": [42, 46, 47, 49, 50], "sec": [20, 22, 73, 74, 75, 77, 79, 80], "second": [1, 3, 6, 9, 10, 12, 20, 21, 23, 24, 26, 70, 79, 82], "secondari": [0, 8, 70, 89], "secondary_offload_min_prior": 70, "secondaryoffloadminprior": 0, "secondli": 79, "section": [3, 6, 16, 17, 19, 20, 27, 28, 30, 65, 69, 71, 73, 75, 77, 78, 79, 80, 82, 88, 91, 93, 97], "section_s": 82, "secur": [45, 93], "securityprotocol": 45, "see": [0, 1, 5, 6, 8, 12, 16, 17, 20, 21, 23, 24, 25, 27, 28, 30, 31, 32, 34, 40, 61, 67, 73, 74, 75, 77, 79, 80, 81, 82, 83, 84, 89, 90, 92, 93, 98], "seed": [0, 6, 30, 38, 39, 70, 86, 93], "seem": [9, 51, 58, 73, 76], "seen": [12, 20, 73], "segment": 93, "select": [0, 4, 6, 18, 25, 26, 28, 29, 73, 80, 82, 87, 89, 96, 99], "selectcontextid": 0, "selectgenidx": 0, "selective_scan": 82, "self": [0, 5, 7, 14, 16, 17, 52, 70, 73, 82, 84, 87, 92, 95, 98, 99], "self_attent": 17, "self_attention_mask": 83, "self_attention_packed_mask": 83, "self_attn": [17, 95], "selfidx": 0, "sell": 73, "semicolon": 65, "senat": [42, 46, 47, 49, 50], "send": [0, 2, 16, 26, 30, 75, 76, 82, 88, 93], "sens": 77, "sensit": [26, 77], "sent": [0, 12, 28, 30, 70], "sentenc": [0, 6, 70, 88], "separ": [11, 12, 29, 54, 65, 73, 82, 87, 97], "separate_match_rewrit": 7, "seq": [1, 5, 73, 82], "seq_idx": 87, "seq_len": [74, 82, 83, 97], "seq_length": 82, "seq_lens_cuda": 97, "seqlen": [0, 82], "seqslot": 1, "sequenc": [0, 1, 3, 5, 6, 7, 8, 9, 12, 16, 20, 21, 22, 23, 24, 26, 27, 28, 70, 71, 73, 74, 75, 78, 81, 82, 83, 87, 89, 93, 97, 98], "sequence_length": [82, 83, 87, 92], "sequence_length_buff": 87, "sequence_limit_length": 87, "sequenceindex": [0, 3], "sequencelengthscba": 1, "sequencelimitlength": 1, "sequenti": [0, 2, 12, 27, 89], "seri": 93, "serial": [29, 82, 84, 87], "serializ": 70, "serialize_engin": 87, "serializeds": 0, "serializedst": 0, "serv": [0, 2, 3, 5, 8, 12, 16, 18, 24, 25, 33, 34, 35, 37, 38, 39, 40, 41, 60, 61, 62, 64, 70, 80, 93, 96, 97], "server": [0, 9, 12, 16, 18, 22, 31, 33, 34, 35, 37, 38, 39, 60, 61, 62, 64, 93], "server_start_timeout": 30, "servic": [18, 59, 64], "session": [5, 69, 73, 87], "set": [0, 1, 2, 3, 4, 5, 6, 7, 8, 11, 12, 13, 15, 17, 18, 19, 20, 26, 28, 29, 30, 36, 45, 55, 56, 57, 65, 67, 70, 71, 72, 74, 75, 77, 79, 80, 81, 82, 83, 84, 85, 87, 88, 89, 92, 93, 99], "set_attn_processor": 84, "set_from_opt": 1, "set_if_not_exist": 84, "set_input_shap": 87, "set_rank": 84, "set_rel_attn_t": 83, "set_shap": 87, "setadditionalmodeloutput": [0, 3], "setallottedtimem": 0, "setbackend": 0, "setbadword": 0, "setbatchingtyp": 0, "setbeamsearchdiversityr": 0, "setbeamwidth": 0, "setbeamwidtharrai": 0, "setbitto": 0, "setcachest": 0, "setcachetransceiverconfig": 0, "setclientid": 0, "setcommst": 0, "setcommunicationmod": 0, "setcommunicationtyp": 0, "setcontextfmha": 1, "setcontextphaseparam": [0, 2], "setcopyonpartialreus": 0, "setcrossattentionmask": 0, "setcrosskvcachefract": 0, "setcudagraphcaches": 0, "setcudagraphmod": 0, "setdatatyp": 1, "setdebugconfig": 0, "setdebuginputtensor": 0, "setdebugoutputtensor": 0, "setdebugtensornam": 0, "setdebugtensorsmaxiter": 0, "setdecodingconfig": 0, "setdecodingmod": 0, "setdeviceid": 0, "seteagleconfig": 0, "seteagleinput": 1, "setearlystop": 0, "setembeddingbia": 0, "setenableblockreus": 0, "setenablechunkedcontext": 0, "setenablecontextfmhafp32acc": 0, "setenablepartialreus": 0, "setenabletrtoverlap": 0, "setencodedvocab": 0, "setencoderhiddens": 1, "setencoderinputfeatur": 0, "setencoderinputtokenid": 0, "setencoderoutputlength": 0, "setendid": 0, "seteventbuffermaxs": 0, "setexecutionconfig": 1, "setexplicitdrafttokensinput": 1, "setextendedruntimeperfknobconfig": 0, "setexternaldrafttokensconfig": 0, "setfreegpumemoryfract": 0, "setfrequencypenalti": 0, "setfrom": 0, "setfrominput": 1, "setgathergenerationlogit": 0, "setgemmallreducedtyp": 1, "setgpuweightsperc": [0, 13], "setguideddecodingconfig": 0, "setguideddecodingparam": 0, "sethostcaches": 0, "setinittozero": 1, "setisorchestr": 0, "setiterstatsmaxiter": 0, "setkvcacheconfig": 0, "setkvcacheretentionconfig": 0, "setkvcachetyp": 1, "setlanguageadapteruid": 0, "setlayertyp": 1, "setlengthpenalti": 0, "setlevel": 1, "setlogitsdtyp": 1, "setlogitspostprocessor": 0, "setlogitspostprocessorconfig": 0, "setlogitspostprocessornam": 0, "setlookaheadconfig": 0, "setlookaheaddecodingconfig": 0, "setloraconfig": 0, "setloramodul": 1, "setmanagedweightsmap": 1, "setmanageweightstyp": 1, "setmaxattentionwindowvec": 0, "setmaxbatchs": [0, 1], "setmaxbeamwidth": [0, 1], "setmaxdraftpathlen": 1, "setmaxdrafttoken": 1, "setmaxencoderlen": 1, "setmaxinputlen": 1, "setmaxlorarank": 1, "setmaxnumpath": 1, "setmaxnumtoken": [0, 1], "setmaxpagesperblock": 1, "setmaxpositionembed": 1, "setmaxpromptembeddingtables": 1, "setmaxqueues": 0, "setmaxseqidlemicrosecond": 0, "setmaxsequencelen": 1, "setmaxtoken": 0, "setmedusachoic": 0, "setmem": 1, "setmemorytyp": 1, "setminp": 0, "setmintoken": 0, "setmlphiddens": 1, "setmodelnam": 1, "setmodelvari": 1, "setmropeconfig": 0, "setmultiblockmod": 0, "setmultimodalembed": 0, "setnbcrosskvhead": 1, "setnbkvhead": 1, "setnorepeatngrams": 0, "setnormalizelogprob": 0, "setnumcopystream": 1, "setnumdecodingenginetoken": 1, "setnumkvheadspercrosslay": 1, "setnumkvheadsperlay": 1, "setnumlanguag": 1, "setnumnod": 0, "setnumreturnsequ": 0, "setonboardblock": 0, "setorchestratorconfig": 0, "setorchleadercomm": 0, "setoutputconfig": 0, "setpadid": 0, "setpagedcontextfmha": 1, "setpagewidth": 1, "setparallelconfig": 0, "setparticipantid": 0, "setpath": 1, "setpeftcacheconfig": 0, "setpositionid": 0, "setppreducescatt": 1, "setpresencepenalti": 0, "setprior": 0, "setprocessorbatch": 0, "setprocessormap": 0, "setprompttableoffload": 0, "setprompttuningconfig": 0, "setquantmod": 1, "setrecvpollperiodm": 0, "setrepetitionpenalti": 0, "setrepl": [0, 3], "setrequeststatsmaxiter": 0, "setrequesttyp": [0, 2], "setreturnallgeneratedtoken": 0, "setrnnconfig": 1, "setrotaryembeddingdim": 1, "setsamplingconfig": 0, "setschedulerconfig": 0, "setse": 0, "setsecondaryoffloadminprior": 0, "setsinktokenlength": 0, "setsizeperhead": 1, "setskipcrossattnblock": [0, 1], "setslotsperpag": 1, "setspawnprocess": 0, "setspecdecconfig": 0, "setspeculativedecodingmod": 1, "setspeculativedecodingmodul": 1, "setstoptokenid": 0, "setstopword": 0, "setstream": 0, "settemperatur": 0, "setter": [0, 6], "settokenizerstr": 0, "settokensperblock": 1, "settopk": 0, "settopp": 0, "settoppdecai": 0, "settoppmin": 0, "settoppresetid": 0, "settotalnumpag": 1, "setup": [1, 5, 29, 45, 55, 56, 57, 67, 76, 77, 87, 88, 89, 93], "setup_fake_prompt": 87, "setup_fake_prompts_qwen2vl": 87, "setup_fake_prompts_vila": 87, "setup_input": 87, "setupeagl": 1, "setupexplicitdrafttoken": 1, "setuplookahead": 1, "setupspeculativedecod": 1, "setuptool": [66, 67], "setusecrossattent": 1, "setusegpudirectstorag": 0, "setusemrop": 1, "setusepositionembed": 1, "setuseshapeinfer": 1, "setusetokentypeembed": 1, "setworkerexecutablepath": 0, "setzero": [0, 1], "seve": 70, "sever": [0, 1, 2, 5, 7, 12, 15, 27, 36, 77, 78, 79, 80, 82, 89, 92, 97], "sft": 58, "sh": [16, 31, 93, 94], "shah": 93, "shaken": 51, "shall": [19, 89], "shape": [0, 1, 5, 7, 10, 15, 16, 26, 28, 70, 80, 82, 84, 87, 89, 90, 92, 93, 97, 98], "shape_cast_dtyp": 82, "shapeequ": 1, "shard": [17, 26, 64, 73, 78, 82, 83], "shard_map": 17, "sharding_along_vocab": 70, "sharding_dim": [82, 83], "share": [1, 2, 3, 5, 7, 8, 9, 10, 12, 19, 20, 25, 26, 27, 28, 29, 65, 76, 77, 82, 83, 93], "share_embed": 93, "share_weight": 83, "shared_embedding_t": 93, "shared_fc1": 28, "shared_fc2": 28, "shared_ptr": [0, 1], "sharedconstptr": 1, "sharedptr": 1, "shelf": 93, "sherlock113": 93, "shift": [11, 27], "ship": [19, 51], "shm": 92, "short": [5, 73, 77, 79], "short_mscal": [82, 83], "shorter": [5, 74], "shot": 93, "should": [0, 1, 2, 3, 7, 9, 10, 11, 19, 20, 28, 36, 42, 45, 46, 47, 49, 50, 52, 53, 55, 56, 57, 58, 65, 70, 73, 74, 75, 76, 80, 81, 82, 83, 85, 87, 89, 93, 95, 97, 98, 99], "should_stop": 87, "shouldus": 5, "show": [2, 3, 16, 22, 26, 27, 28, 30, 40, 74, 75, 79, 80, 88, 89, 91, 94], "showcas": [77, 80, 88], "shown": [11, 23, 27, 30, 65, 69, 73, 75, 77, 79, 80, 82], "shrunk": 82, "shuffl": 82, "shut": 2, "shutdown": [0, 59, 69, 70], "si": 5, "sibl": 16, "side": [3, 82], "side_stream_id": 82, "sidestreamidtyp": 82, "sigh": 58, "sigmoid": [16, 82], "signal": 0, "signatur": [7, 52, 82], "signifi": 79, "signific": [3, 5, 8, 23, 27, 28, 58, 76, 77, 79, 80], "significantli": [25, 26, 27, 28, 75, 76, 77, 79, 80, 89, 97], "silicon": 28, "silu": [16, 82, 83], "similar": [0, 5, 6, 7, 12, 20, 21, 23, 27, 36, 49, 53, 72, 73, 81, 82, 96, 99], "similarli": 12, "simpl": [2, 7, 8, 12, 16, 40, 52, 65, 69, 71, 74, 88, 94], "simpler": 12, "simpleschedul": 99, "simplest": 82, "simpli": [5, 12, 71, 73, 74, 79, 88, 92, 95], "simplic": 19, "simplifi": [5, 19, 73, 79, 82, 93], "simultan": [12, 79], "sin": [0, 82, 83], "sinc": [0, 1, 4, 5, 7, 9, 12, 13, 19, 20, 27, 28, 31, 36, 65, 70, 73, 75, 76, 77, 79, 80, 82, 84, 89, 96, 98, 99], "sincer": 28, "sinco": 83, "singl": [0, 1, 2, 3, 4, 5, 6, 8, 12, 14, 16, 19, 20, 23, 24, 26, 27, 28, 29, 34, 52, 61, 69, 70, 72, 73, 77, 80, 82, 84, 88, 89, 90, 93, 95, 96, 97, 98], "singleton": [7, 82], "sink": [0, 1, 5, 70, 87], "sink_token_len": 87, "sink_token_length": [5, 70, 87], "sinktokenlength": [0, 1], "sinusoid": 83, "sit": [19, 58], "situaiton": 74, "situat": [12, 58, 64, 75, 79], "six": 27, "size": [0, 1, 2, 5, 6, 8, 9, 10, 11, 12, 13, 20, 22, 23, 25, 26, 27, 28, 29, 30, 36, 52, 55, 56, 57, 64, 70, 72, 73, 74, 75, 76, 77, 78, 80, 82, 83, 84, 87, 92, 93, 97, 99], "size_t": [0, 1], "size_typ": [0, 1], "sizeof": 1, "sizeperhead": [0, 1], "sizetype32": [0, 1], "sizetype64": [0, 1], "skip": [0, 1, 7, 17, 20, 32, 59, 65, 70, 82, 99], "skip_attn": [82, 83], "skip_cross_attn_block": [84, 87], "skip_cross_kv": [83, 87], "skip_encod": 87, "skip_special_token": [70, 93], "skip_tokenizer_init": [36, 70], "skipcrossattnblock": [0, 1], "sku": [75, 77, 79, 80], "skywork": [90, 91, 93], "sleep": 32, "slice": [1, 4, 17, 82, 93], "slice_shap": 17, "sliceinputtyp": 82, "slicen": 1, "slide": [8, 64, 81, 82, 87, 93], "slider": [20, 26, 73], "sliding_window": 84, "sliding_window_caus": 82, "sliding_window_pattern": 84, "slight": [20, 27, 28, 77, 79, 80], "slightli": [0, 2, 10, 11, 30, 77, 80], "slope": [5, 82], "slot": [0, 1, 93], "slot_map": [82, 84], "slotidx": 1, "slotsperpag": 1, "slow": [3, 9, 70, 71, 76], "slower": [8, 19, 28, 76], "slowest": 5, "slurm": [16, 55, 56, 57, 67, 69, 92, 93], "sm": [91, 93], "sm120": 93, "sm80": [91, 93], "sm86": [91, 93], "sm89": [91, 93], "sm90": [91, 93], "small": [5, 9, 11, 12, 16, 25, 26, 27, 28, 75, 77, 79, 80, 82, 89, 92, 93], "smaller": [1, 12, 20, 27, 29, 72, 73, 76, 79, 80, 81, 82, 89, 93], "smallest": [0, 1, 8, 82], "smart": 82, "smaug": [91, 93], "smi": [20, 26, 73, 89], "smile": 58, "smith": [42, 46, 47, 48, 49, 50, 52, 59], "smooth": [19, 70, 93], "smoother": 20, "smoothquant": [7, 25, 64, 93], "smoothquant_v": 70, "snapshot": 73, "snapshot_download": 58, "snip": 73, "snippet": [73, 93, 99], "snshrivas10": 58, "so": [0, 2, 3, 5, 7, 10, 12, 18, 19, 20, 26, 27, 28, 31, 36, 49, 65, 70, 73, 76, 77, 79, 80, 81, 82, 83, 84, 89, 91, 93, 95, 98], "socketst": 0, "softmax": [5, 16, 27, 28, 82, 97], "softplu": 82, "softwar": [3, 5, 16, 28, 64, 71, 93], "solid": 78, "solut": [18, 69, 92, 96], "some": [0, 2, 3, 4, 5, 6, 7, 9, 12, 13, 15, 16, 19, 20, 26, 27, 28, 29, 30, 32, 58, 67, 70, 71, 74, 77, 78, 80, 81, 82, 85, 88, 89, 92, 93, 95, 96, 99], "someth": [16, 36, 51], "sometim": 73, "song": 73, "soon": [0, 21, 22, 23, 24, 25, 27, 36], "sophist": 52, "sora": [34, 61], "sort": [0, 1, 3, 6, 82], "sota": 93, "sourc": [14, 15, 17, 19, 20, 21, 24, 26, 28, 29, 30, 33, 34, 35, 37, 38, 39, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 64, 70, 71, 82, 83, 84, 85, 86, 87, 93], "source_root": [55, 56, 57], "sourcetaskvalu": 1, "soyer": [14, 16, 92], "space": [10, 65, 70, 79, 89, 98], "spaces_between_special_token": [70, 93], "span": [19, 26, 27], "spars": [12, 28, 82, 93], "sparse_fc1": 28, "sparse_fc2": 28, "sparsiti": 29, "spatial_norm_dim": 83, "spawn": [40, 50, 66, 67, 69, 75, 88, 92], "spawnprocess": [0, 2], "spec": 29, "spec_decode_algo": 27, "spec_decode_nextn": 27, "spec_decoding_generation_length": [82, 83, 84], "spec_decoding_is_generation_length_vari": [82, 83, 84], "spec_decoding_max_generation_length": [82, 83], "spec_decoding_packed_mask": [82, 83, 84], "spec_decoding_param": [83, 84], "spec_decoding_position_offset": [82, 83, 84], "spec_decoding_us": [82, 83], "specdec": 0, "specdecconfig": 0, "specdecfastlogitsinfo": 0, "specdecodinggenerationlength": 1, "specdecodinggenerationlengthshost": 1, "specdecodingpackedmask": 1, "specdecodingparam": 83, "specdecodingpositionoffset": 1, "specdecodingstat": 0, "specdecstat": 0, "special": [2, 5, 10, 16, 17, 21, 27, 29, 70, 93], "specif": [0, 1, 4, 6, 7, 8, 10, 11, 12, 15, 19, 22, 25, 26, 28, 30, 52, 65, 67, 73, 76, 77, 80, 82, 88, 93, 95, 96], "specifi": [0, 1, 2, 3, 5, 6, 7, 8, 10, 12, 17, 19, 20, 29, 30, 36, 43, 44, 45, 52, 54, 58, 59, 65, 69, 70, 72, 73, 74, 76, 77, 79, 81, 82, 84, 85, 87, 88, 89, 92, 93, 97], "specul": [0, 1, 3, 26, 64, 68, 70, 73, 75, 82, 93], "speculative_config": [20, 26, 27, 43, 44, 53, 54, 70], "speculative_decod": 93, "speculative_decoding_draft_tokens_extern": 84, "speculative_decoding_mod": [29, 70, 73], "speculative_model": [43, 44, 54, 70], "speculativedecod": 0, "speculativedecodingconfig": 0, "speculativedecodingfastlogitsinfo": 0, "speculativedecodingmetr": 0, "speculativedecodingmod": [70, 84, 93], "speculativedecodingmodul": 93, "speculativedecodingoutput": 1, "speed": [16, 22, 26, 27, 28, 29, 73, 74, 80, 93], "speedup": [22, 24, 25, 26, 28], "spent": 0, "split": [1, 4, 5, 10, 16, 70, 73, 76, 77, 82, 89, 93], "split_input_id": 87, "split_prompt_by_imag": 87, "split_siz": 82, "split_size_or_sect": 82, "splittransposecpu": 1, "splittransposecpuinn": 1, "splitwis": 2, "spot": 79, "sq": [25, 90, 93], "sqrt": [5, 82], "squar": [79, 82], "squared_relu": 82, "squeez": [1, 82, 87], "src": [1, 16, 82], "src_seq_len": 82, "srcdesc": 0, "srctype": 1, "srun": [16, 30, 55, 56, 57, 67, 92], "sshd": 31, "ssid": 45, "ssm": 82, "ssm_state": 84, "stabil": 26, "stabl": [5, 17, 29, 75, 79, 80, 82, 93], "stack": [17, 26, 65, 82], "stage": [0, 5, 7, 12, 27, 74, 89, 93, 97], "stai": [22, 25, 76, 80], "stand": 16, "standalon": 19, "standard": [12, 16, 18, 21, 74, 82], "starcod": [69, 91, 93], "starcoder1": 90, "starcoder2": [90, 93], "starrickliu": 93, "start": [0, 3, 5, 7, 9, 20, 27, 29, 31, 32, 33, 34, 35, 37, 38, 39, 57, 58, 60, 61, 62, 65, 69, 70, 71, 73, 74, 75, 76, 79, 81, 82, 84, 86, 87, 89, 93], "start_dim": 82, "startup": 92, "stat": [0, 70, 93], "state": [0, 1, 3, 4, 5, 7, 8, 9, 12, 20, 26, 27, 29, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 59, 66, 67, 70, 73, 74, 75, 79, 81, 82, 88, 93, 94, 99], "state_dtyp": 87, "state_or_ptr": 82, "state_s": 87, "statement": 69, "stateptr": 0, "states": 1, "static": [0, 1, 3, 12, 28, 29, 70, 82, 83, 84, 87, 93], "static_batch": [70, 81], "static_cast": 90, "staticbatchingstat": 0, "statist": [0, 3, 12, 30, 70, 73, 93], "statu": 92, "std": [0, 1, 3], "stddev": [30, 38, 39], "stdev": [20, 56, 72, 73, 74, 75], "stdit": 93, "stdout": [20, 56, 72, 73, 74, 75], "steadi": 74, "steady_clock": 0, "step": [0, 1, 5, 6, 7, 9, 12, 15, 16, 18, 19, 21, 26, 27, 32, 52, 64, 66, 67, 70, 71, 73, 74, 75, 82, 87, 92, 96, 97, 98, 99], "still": [5, 17, 19, 20, 26, 27, 28, 71, 73, 75, 77, 82, 87, 89, 93], "stop": [0, 1, 3, 6, 7, 12, 70, 73, 79, 87, 88, 93], "stop_reason": [53, 70, 88, 93], "stop_token_id": [3, 70], "stop_words_data": 87, "stop_words_list": 87, "stopping_criteria": 87, "stoppingcriteria": [87, 93], "stoppingcriterialist": 87, "stoptokenid": [0, 3], "stopword": 0, "stopwordslen": 1, "stopwordslist": 1, "stopwordsptr": 1, "storag": [0, 8, 10, 70], "store": [0, 1, 5, 8, 9, 10, 16, 22, 26, 51, 54, 69, 70, 73, 81, 82, 84, 89, 90, 95, 97, 98], "store_tru": 54, "stored_block": 51, "stori": 58, "str": [15, 19, 47, 48, 70, 82, 83, 84, 87], "straightforward": 27, "strategi": [0, 11, 12, 25, 27, 36, 49, 64, 70, 73, 78, 82, 84, 89, 93], "stream": [0, 1, 2, 3, 16, 28, 29, 30, 36, 38, 39, 40, 41, 52, 70, 72, 82, 87, 89, 92, 93], "stream_ptr": 52, "streaming_llm": 93, "streamingllm": [29, 64, 93], "streamlin": [73, 88], "streamptr": [0, 1, 3], "street": 58, "strenum": [70, 86], "strict": [26, 27], "strict_bound": 82, "strict_dtyp": [82, 83], "stricter": 26, "strictli": 73, "stride": [1, 82, 83], "strike": [12, 51], "string": [0, 1, 3, 15, 45, 70, 73, 82, 87], "string_valu": 9, "string_view": 1, "stringptrmap": 1, "stringvec": 0, "strip": [29, 93], "strip_plan": 29, "strongli": 77, "strongly_typ": [70, 93], "struct": [0, 1, 8], "structur": [0, 4, 7, 8, 12, 28, 52, 70, 82, 89, 93], "structural_tag": 70, "struggl": 58, "student": [42, 46, 47, 49, 50, 52], "studi": [28, 75, 77, 78, 80], "style": [5, 12, 26, 93], "sub": [15, 19, 82], "subclass": [1, 19, 52, 95], "subcommad": 73, "subcommand": [74, 93], "subgraph": [7, 82], "subject": [2, 21, 23, 24, 25, 69, 82, 88, 94], "submiss": 73, "submit": [10, 70, 73], "submit_sync": 70, "submittransferrequest": 0, "submodul": [20, 65, 95], "suboptim": 16, "subscript": 82, "subsequ": [2, 9, 10, 12, 27, 75], "subset": [0, 3, 6, 16, 19, 27, 73, 82], "substanti": [9, 12, 26, 28], "subsystem": 93, "subtract": 7, "succe": [89, 93], "succeed": 87, "success": [3, 22, 26, 74], "successfulli": [12, 32, 77], "sudo": [20, 26, 66, 67, 73], "suffer": 26, "suffici": [76, 77], "suggest": [5, 25, 58, 77], "suit": [5, 73, 74], "sum": [1, 7, 14, 82, 98], "sum_of_token": 82, "summar": [5, 12, 13, 14, 15, 23, 25, 73, 74, 81, 89], "summari": [8, 12, 64], "summat": 82, "sunjiabin17": 93, "super": [7, 14, 17, 19, 91, 92, 95, 99], "superchip": 91, "supplementari": 83, "suppli": [10, 18], "support": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 15, 18, 19, 21, 22, 23, 24, 25, 26, 28, 29, 30, 31, 36, 45, 52, 55, 56, 57, 58, 64, 67, 68, 70, 74, 75, 77, 79, 80, 81, 82, 83, 85, 88, 92, 93, 94, 95, 96, 97, 98, 99], "supportsinflightbatch": 1, "suppos": 95, "suprem": [42, 46, 47, 49, 50], "sure": [2, 19, 20, 27, 32, 65, 73, 81, 82, 93], "surpass": 5, "surround": [5, 93], "swa": 8, "swap": 8, "sweep": [16, 22, 79], "sweet": 79, "swept": 23, "swiglu": [29, 82, 93], "switch": [4, 9, 11, 22, 25, 26, 28, 65, 81, 89, 93], "sxm": [22, 29, 75, 77, 78], "sy": 93, "symbol": 0, "sync": 87, "synchron": [1, 3, 16, 70, 92, 93], "syncmessag": 0, "syntax": [82, 88], "synthet": [20, 30, 38, 39, 73, 74], "synthetic_128_128": 73, "synthetic_2048_2048": 75, "synthetic_2048_2048_1000": 75, "system": [8, 9, 16, 20, 22, 27, 28, 30, 33, 34, 45, 55, 56, 57, 60, 61, 64, 65, 67, 74, 76, 88, 91, 93, 94], "systemat": 26, "t": [0, 1, 5, 12, 16, 19, 26, 28, 30, 31, 36, 51, 55, 56, 57, 67, 70, 72, 73, 76, 79, 80, 82, 84, 87, 92], "t5": [5, 6, 90, 91, 93], "t_": 27, "t_2": 27, "t_5": 27, "tabl": [0, 6, 9, 22, 25, 29, 73, 74, 82, 83, 87, 91, 92, 93], "tackl": 28, "tactic": [28, 29], "tag": [0, 31, 65, 70], "tailor": [25, 77, 80], "take": [0, 1, 2, 5, 6, 7, 9, 11, 15, 19, 27, 51, 58, 71, 73, 75, 76, 79, 82, 83, 98], "taken": [17, 21, 22, 82], "talk": 58, "tanh": [82, 83], "target": [0, 17, 20, 28, 29, 36, 64, 65, 73, 80, 81, 93], "target_isl": 73, "target_osl": 73, "targetcach": 1, "targetpageid": 1, "targetprob": 1, "targettaskvalu": 1, "tarot": 58, "task": [0, 1, 9, 10, 12, 14, 15, 47, 48, 55, 56, 57, 70, 73, 83, 87, 90, 93, 98], "task_id": [10, 73], "task_vocab_s": 83, "taskid": [0, 1], "taskidtyp": 1, "tasklayermoduleconfig": 1, "tasklayermoduleconfigbind": 1, "tasklayermoduleconfiglistptr": 1, "taskshost": 1, "taskvalu": 1, "taskvalueptr": 1, "taslid": 1, "tayef": 93, "tconstptr": 1, "tcp": 32, "team": [15, 19, 20, 26, 27, 28, 32, 91, 93], "tech": [27, 93], "technic": [8, 27, 28, 64], "techniqu": [5, 7, 12, 16, 21, 26, 27, 28, 71, 76, 77, 78, 81, 90, 93], "technologi": [26, 42, 46, 47, 49, 50, 52], "tekit_2025": 73, "tell": [34, 58, 59, 61, 80, 88], "temb": 83, "temp": 87, "temperatur": [0, 1, 6, 30, 33, 34, 35, 36, 40, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 54, 59, 66, 67, 70, 73, 75, 81, 87, 88, 93], "tempfil": [46, 49], "templat": [0, 1, 16, 17], "tempor": 87, "temporari": 2, "ten": [12, 25, 27], "tend": 81, "tensor": [1, 6, 11, 15, 16, 17, 20, 21, 22, 23, 24, 26, 27, 28, 30, 50, 52, 64, 70, 73, 74, 77, 78, 80, 82, 83, 84, 87, 90, 92, 93, 95, 97], "tensor_dict": 87, "tensor_input": 7, "tensor_parallel_s": [50, 51, 54, 55, 56, 57, 70, 75, 76, 77, 80, 81], "tensor_shap": 17, "tensorconstptr": 1, "tensorinfo": 87, "tensorloc": 82, "tensormap": 1, "tensorparallel": [0, 1, 6], "tensorptr": [0, 1], "tensorrt": [1, 3, 5, 6, 7, 8, 11, 13, 14, 21, 24, 26, 28, 29, 30, 33, 34, 35, 36, 37, 38, 39, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 66, 67, 68, 72, 74, 77, 78, 80, 81, 82, 87, 90, 92, 94, 95, 96, 97, 98, 99], "tensorrt_llm": [0, 1, 2, 3, 5, 6, 7, 10, 13, 14, 16, 17, 19, 20, 30, 31, 32, 36, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 65, 66, 67, 70, 73, 74, 75, 77, 80, 81, 82, 83, 84, 85, 86, 87, 88, 92, 93, 94, 95, 96, 97, 98], "tensorrt_llm_gpt": 16, "tensorrt_llm_rouge1_threshold": 15, "tensorrtllm_backend": [10, 88, 93], "term": [16, 69, 81, 82, 88], "termin": [0, 9, 32, 74, 93], "test": [5, 25, 26, 27, 30, 34, 61, 64, 65, 66, 67, 73, 74, 75, 77, 78, 79, 80, 81, 91, 93, 98], "test_graph_rewrit": 7, "test_trt_llm": [13, 14, 15], "texec": 0, "text": [0, 3, 5, 6, 9, 29, 34, 36, 40, 41, 42, 50, 51, 59, 61, 66, 67, 70, 71, 73, 74, 75, 81, 87, 88, 91, 92, 93, 94], "text_diff": 70, "text_hidden_s": 84, "textattack": 91, "textprompt": 70, "tg_group": 82, "tgt": [16, 82], "tgt_len": [82, 83], "tgt_seq_len": 82, "th": [1, 15, 27, 82], "than": [0, 1, 2, 3, 5, 6, 7, 9, 12, 16, 20, 21, 22, 23, 25, 26, 27, 28, 29, 65, 70, 71, 73, 74, 75, 76, 77, 79, 81, 82, 87, 89, 92, 93, 97], "thank": [27, 93], "thecodewrangl": 93, "thei": [0, 1, 3, 5, 6, 10, 16, 17, 19, 26, 27, 28, 53, 65, 70, 73, 75, 77, 79, 80, 81, 82, 84, 90, 93], "them": [0, 3, 4, 7, 12, 13, 20, 26, 27, 28, 55, 56, 57, 70, 71, 72, 73, 76, 78, 79, 81, 82, 87, 89, 95], "theoret": 89, "theori": 81, "therebi": [2, 81], "therefor": [13, 19, 74, 82, 92, 98], "thermal": 73, "theta": 82, "thi": [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 23, 25, 26, 27, 28, 29, 30, 31, 32, 36, 40, 45, 52, 54, 55, 56, 57, 58, 65, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 87, 88, 89, 90, 92, 93, 94, 95, 96, 97, 98, 99], "thin": 19, "thing": [6, 32, 42, 46, 47, 49, 50, 52, 79, 80], "think": [26, 27, 28, 51, 78], "third": [3, 93], "those": [3, 5, 6, 15, 16, 18, 20, 26, 27, 28, 29, 30, 72, 74, 75, 80, 82, 83, 90], "though": [19, 27, 79, 89], "thread": [0, 1, 5, 11, 36, 69, 73, 87], "three": [2, 3, 15, 25, 26, 28, 81, 82, 90, 95, 96, 97], "threshold": [0, 26, 27, 82, 87], "throttl": 73, "through": [0, 5, 6, 7, 11, 12, 16, 17, 18, 20, 26, 29, 30, 65, 71, 73, 75, 76, 77, 79, 80, 83, 88, 93], "throughout": [75, 78], "throughput": [0, 3, 5, 21, 22, 23, 27, 56, 64, 72, 77, 79, 80, 81, 93, 97], "throw": [0, 1], "thu": [9, 19, 20, 26, 28, 65, 82, 89], "thumb": [5, 76, 92], "ti": [5, 27], "tiiuae": 73, "tile": 28, "time": [0, 1, 2, 3, 5, 9, 10, 11, 12, 13, 16, 20, 23, 25, 26, 27, 28, 29, 42, 46, 47, 48, 49, 50, 58, 64, 65, 70, 71, 72, 73, 74, 75, 77, 78, 79, 81, 82, 87, 92, 93, 98], "time_embed_dim": 83, "time_encod": 87, "time_point": 0, "timedelta": 70, "timedout": 0, "timelin": 15, "timeout": [0, 30, 36, 70, 93], "timepoint": 0, "timestamp": 0, "timestep": [83, 84], "timestepembed": 83, "timingmetr": 0, "tini": 58, "tinyllama": [30, 33, 35, 38, 40, 42, 45, 46, 47, 48, 49, 50, 51, 52, 53, 58, 59, 60, 62, 66, 67, 69, 88, 94], "tip": 64, "titl": 45, "tle": 13, "tllm_checkpoint_16gpu_tp8_pp2": 76, "tllm_ckpt_dir": 14, "tllm_engine_dir": 14, "tllm_kei": [17, 83], "tllm_llmapi_build_cach": 93, "tllm_llmapi_enable_nvtx": 72, "tllm_log_level": 92, "tllm_nvtx_debug": 72, "tllm_override_layer_num": 93, "tllm_profile_record_gc": 72, "tllm_profile_start_stop": 72, "tllm_to_externel_key_dict": 17, "tllm_torch_profile_trac": 72, "tllm_trace_model_forward": 93, "tllm_weight": 17, "tllmruntim": [1, 6, 92], "tlntin": 93, "tmp": [10, 13, 56, 72, 73, 76], "tmp9so41y3r": 73, "tmpowsrb_f4": 73, "tmpxhdvasex": 73, "to_arrai": 82, "to_dict": [70, 84], "to_json_fil": 84, "to_layer_quant_config": 84, "to_legacy_set": 85, "to_str": [0, 1, 3], "to_trt": 84, "tobyt": 1, "todo": [1, 54, 82], "togeth": [3, 5, 6, 10, 16, 18, 21, 26, 27, 29, 87, 90, 93], "toggl": 72, "toi": 79, "toitensor": 0, "tojsonstr": 0, "tok": [21, 23, 24, 80], "token": [0, 1, 2, 3, 4, 5, 6, 8, 9, 12, 16, 20, 21, 24, 25, 26, 27, 28, 29, 30, 31, 38, 39, 45, 51, 52, 56, 64, 70, 72, 73, 74, 75, 77, 78, 80, 82, 83, 84, 87, 88, 89, 90, 93, 95, 96, 97], "token_drop": 83, "token_end": 70, "token_extra_id": 51, "token_id": [36, 51, 52, 53, 70], "token_ids_diff": 70, "token_range_retention_config": 70, "token_start": 70, "token_type_id": [84, 87], "tokenend": 0, "tokenextraid": 1, "tokenextraidtyp": 1, "tokenid": 1, "tokenidtyp": [0, 1], "tokenization_utils_bas": 70, "tokenizer_dir": [14, 16, 88, 92], "tokenizer_image_token": 87, "tokenizer_max_seq_length": [70, 77, 84, 86], "tokenizer_mod": 70, "tokenizer_revis": 70, "tokenizer_str": [0, 3], "tokenizerbas": 70, "tokenizerstr": [0, 3], "tokenlogprob": 70, "tokenrangeretentionconfig": [0, 70], "tokenrangeretentionprior": 0, "tokens_per_block": [8, 9, 29, 87, 93, 98], "tokensperblock": [0, 1, 6], "tokensperstep": 1, "tokensprompt": 70, "tokenstart": 0, "tokyo": [34, 61], "toler": 25, "tomodulenam": 1, "tomoduletyp": 1, "tonylek": 93, "too": [3, 5, 20, 28, 75, 79, 92], "took": 75, "tool": [2, 15, 20, 28, 64, 69, 73, 93], "tool_cal": 88, "toolkit": [18, 19, 25, 26, 67, 96], "top": [0, 5, 6, 12, 16, 18, 27, 28, 70, 82, 93], "top1": 26, "top_k": [6, 70, 87, 93], "top_p": [6, 40, 42, 43, 44, 46, 47, 48, 49, 50, 51, 52, 54, 59, 66, 67, 70, 75, 81, 87, 88], "top_p_decai": [70, 87], "top_p_min": [70, 87], "top_p_reset_id": [70, 87], "topenkoff": 93, "topic": 80, "topk": [0, 1, 4, 6, 12, 26, 28, 82, 93], "topk_logit": 3, "topklastdim": 82, "topklogit": 3, "topkmedusahead": 1, "topktopp": [0, 6], "topmodelmixin": [19, 84], "topn": 26, "topp": [0, 1, 6, 93], "toppdecai": [0, 1, 6], "toppmin": [0, 1, 6, 70], "toppresetid": [0, 1, 6], "torch": [5, 17, 52, 59, 65, 66, 67, 70, 73, 82, 87, 92, 95], "torch_compile_en": 70, "torch_compile_enable_userbuff": 70, "torch_compile_fullgraph": 70, "torch_compile_inductor_en": 70, "torch_compile_piecewise_cuda_graph": 70, "torchaudio": [66, 67], "torchllmarg": 70, "torchvis": [66, 67], "tostr": [0, 1], "total": [0, 1, 4, 5, 6, 12, 15, 17, 20, 27, 29, 30, 73, 74, 75, 76, 89, 98], "total_lat": [21, 24], "total_token": 88, "totalaccepteddrafttoken": 0, "totaldrafttoken": 0, "totalgentoken": 1, "totalnumpag": 1, "totensor": 0, "touch": [31, 95], "tp": [0, 2, 4, 6, 10, 16, 20, 21, 22, 23, 24, 25, 26, 27, 28, 30, 56, 73, 74, 75, 82, 93], "tp1": [21, 22, 23], "tp2": 73, "tp4": 26, "tp4ep2": 26, "tp8": [23, 26, 28], "tp8ep2": 26, "tp_1_pp_1": 73, "tp_dim": [17, 83], "tp_group": [82, 83], "tp_rank": [17, 82, 83], "tp_size": [4, 10, 15, 16, 17, 19, 30, 37, 55, 57, 73, 74, 76, 82, 83, 86, 93], "tp_split_dim": 83, "tpot": [24, 74], "tprank": 1, "tpsize": 1, "tqdm": [17, 70, 93], "trace": [19, 29, 30, 72, 92], "track": [5, 8, 70, 82], "trade": [9, 28], "tradeoff": [25, 26, 27, 77], "tradit": 0, "train": [12, 14, 15, 16, 18, 19, 22, 25, 27, 73, 82, 92, 95], "trait": 93, "transa": 82, "transb": 82, "transceiv": [0, 70], "transfer": [0, 2, 16, 28, 52, 70, 93], "transfer_mod": 70, "transferdesc": 0, "transfermod": 0, "transferop": 0, "transferrequest": 0, "transferstatu": 0, "transform": [0, 4, 5, 12, 14, 15, 16, 17, 29, 30, 36, 70, 84, 88, 89, 91, 92, 93, 95, 96, 98], "translat": [81, 93], "transmiss": [2, 11], "transmit": [2, 11], "transpos": [1, 15, 82], "transposit": 82, "travers": 16, "treat": [5, 26, 82], "tree": [0, 73, 87, 92, 98], "tri": [28, 99], "tricki": 84, "trigger": [5, 7, 16, 29, 36, 59, 69], "trim": 1, "trimpool": 1, "triton": [9, 10, 12, 16, 18, 64, 71, 93], "tritonserv": 93, "trivial": 16, "troubleshoot": [64, 93], "trt": [0, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 16, 17, 22, 31, 46, 49, 73, 79, 82, 84, 86, 87, 89, 92, 93, 97], "trt_ckpt": [10, 13, 15, 92], "trt_engin": [10, 13, 15, 92], "trt_root": 20, "trt_tensor": [16, 82], "trtdatatyp": 1, "trtgptmodel": 89, "trtgptmodeloptionalparam": 93, "trtgptmodelv1": 93, "trtllm": [9, 10, 13, 14, 15, 16, 19, 20, 27, 33, 34, 35, 36, 37, 38, 39, 40, 41, 55, 60, 61, 62, 64, 69, 70, 73, 74, 77, 78, 79, 80, 89, 92, 93], "trtllm_dg_jit_use_nvcc": 20, "trtllm_disable_kv_cache_transfer_overlap": 2, "trtllm_disable_unified_convert": 17, "trtllm_enable_kvcache_receive_parallel": 2, "trtllm_enable_mmha_multi_block_debug": 73, "trtllm_enable_pdl": [20, 26, 27, 73], "trtllm_force_xqa": 5, "trtllm_kvcache_send_max_concurrency_num": 2, "trtllm_kvcache_transfer_buffer_s": 2, "trtllm_kvcache_transfer_use_async_buff": 2, "trtllm_mmha_blocks_per_sequ": 73, "trtllm_mmha_kernel_block_s": 73, "trtllm_model": 17, "trtllm_modules_to_hf_modul": 87, "trtllm_parallel_cache_send": 2, "trtllm_pdl_overlap_ratio": 73, "trtllm_precompiled_loc": 65, "trtllm_prefetch_ratio": 73, "trtllm_request_kv_cache_concurr": 2, "trtllm_serv": 30, "trtllm_try_zcopy_for_kvcache_transf": 2, "trtllm_use_mpi_kvcach": 2, "trtllm_use_precompil": 65, "trtllm_use_ucx_kvcach": 2, "trtllmarg": 70, "trtllmattent": 97, "trtlmmdatatyp": 0, "true": [0, 1, 3, 6, 7, 9, 12, 15, 20, 26, 27, 28, 30, 36, 42, 43, 44, 48, 49, 51, 52, 53, 54, 56, 58, 70, 72, 73, 74, 77, 80, 82, 83, 84, 85, 87, 89, 92, 93], "true_output_valu": 82, "true_valu": 82, "truncat": [70, 93], "truncate_prompt_token": [70, 93], "trust": [28, 70], "trust_remote_cod": [30, 70, 93], "try": [0, 1, 3, 14, 19, 53, 58, 69, 74, 77, 79, 80, 81, 88, 89, 92, 94], "tsuji": 73, "ttensor": 1, "ttft": [74, 77, 79, 80, 81, 93], "ttim": 93, "ttl": 26, "tunabl": 78, "tune": [0, 2, 3, 12, 22, 25, 26, 28, 29, 64, 70, 73, 74, 77, 80, 83, 84, 87, 88, 89, 93], "tuner": 0, "tupl": [0, 1, 82, 83, 87, 99], "turn": [5, 6, 9, 12, 28, 65, 77, 87, 89, 93], "tushar": 93, "tweak": 81, "twice": 16, "two": [0, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 15, 16, 19, 22, 26, 27, 28, 29, 30, 34, 61, 65, 69, 73, 75, 77, 79, 81, 82, 83, 85, 93, 96, 98, 99], "twofold": 12, "twoshot": 82, "txt": [19, 20, 56, 67, 72, 73, 75, 88, 93], "type": [1, 2, 3, 5, 6, 7, 10, 15, 16, 22, 25, 28, 29, 30, 33, 34, 35, 38, 39, 45, 51, 52, 54, 61, 70, 73, 77, 80, 82, 84, 86, 87, 88, 90, 91, 92, 93, 95, 96, 97, 98], "typedef": [0, 1], "typenam": [0, 1, 16], "typetrait": 0, "typic": [0, 2, 7, 14, 16, 19, 25, 27, 28, 30, 67, 69, 76, 77, 80, 81, 85, 87, 89, 93, 95], "typo": 93, "u": [1, 7, 28, 31, 42, 46, 47, 48, 49, 50, 59, 73, 74, 93], "ub": 82, "ub_oneshot": 73, "ub_tp_siz": 73, "ubuntu": [66, 67, 93, 94], "uc_handl": 1, "uc_ptr": 1, "uc_va": 1, "ucx": [2, 93], "ucx_cuda_copy_async_mem_typ": 2, "ucx_cuda_copy_dmabuf": 2, "ucx_info": 2, "ucx_memtype_cach": 2, "ucx_rndv_frag_mem_typ": 2, "ucx_rndv_pipeline_error_handl": 2, "uid": [0, 87], "uint16_t": 0, "uint32": 1, "uint32_t": [0, 1, 82], "uint64": [1, 9], "uint64_t": [0, 1], "uint8": 1, "uint8_t": [0, 1], "uintptr_t": [0, 1], "uk": 28, "uk_bgemm": 26, "ulimit": [65, 92], "ultim": 76, "ulyss": 93, "unabl": [67, 79], "unaccept": 77, "unari": 82, "unaryoper": 82, "unbind": 82, "uncas": 91, "uncertainti": 12, "unchang": [12, 80, 82], "uncommon": 16, "undefin": 82, "under": [0, 25, 29, 65, 69, 73, 74, 92, 93], "underli": [0, 1, 7, 12], "underlying_type_t": 1, "underlyingtyp": [0, 1], "underscor": 77, "understand": [64, 65, 72], "understood": [70, 79], "underutil": 12, "uneven": 93, "unevenli": 26, "unexpect": [92, 93], "unfinish": 0, "unfus": 82, "unfuse_qkv_project": 84, "ungath": 1, "unguid": 45, "unif": 93, "unifi": [15, 19, 25, 93], "uniform": [73, 74, 82], "uniniti": 97, "uninstal": 67, "union": [70, 82], "uniqu": [0, 5, 6, 8, 10, 12, 15, 29, 70, 73], "unique_ptr": [0, 1], "unique_token": 51, "uniqueconstptr": 1, "uniqueptr": 1, "uniquetoken": 1, "unit": [1, 8, 17, 28, 40, 42, 43, 44, 46, 47, 48, 49, 50, 52, 54, 59, 64, 65, 66, 67, 73, 75, 81, 88, 94], "univers": [42, 46, 47, 49, 50, 52], "unless": [0, 36, 70, 76, 80, 81], "unlik": [9, 12], "unlock": 71, "unnecessari": [7, 93, 95, 99], "unneed": [5, 26], "unordered_map": [0, 1, 3], "unpatchifi": 84, "unschedul": 79, "unset": 81, "unsign": 1, "unspecifi": [29, 30, 82], "unsqueez": [1, 82], "unstabl": 19, "unsupport": 93, "until": [0, 1, 3, 6, 9, 12], "untouch": 82, "unus": [0, 73], "up": [0, 5, 6, 10, 12, 20, 22, 23, 26, 27, 28, 29, 45, 70, 73, 79, 80, 93, 98], "up_proj": 17, "upcast": 82, "upcast_attent": 83, "upcast_softmax": 83, "upcom": [25, 98], "updat": [0, 8, 12, 16, 17, 19, 20, 23, 27, 28, 29, 31, 52, 65, 70, 82, 87, 92, 98], "update_from_dict": 70, "update_key_map": 17, "update_kv_cache_typ": 70, "update_output_ids_by_offset": 87, "update_resourc": [96, 98], "update_strategi": 82, "updatenumreturnbeam": 0, "updatespositionid": 1, "upgrad": [66, 67, 88], "uplift": [77, 79, 80], "upon": [12, 74, 80, 92, 93], "upper": [73, 82, 89], "uq_qr_gemm": 26, "url": [30, 34, 38, 39, 61, 65, 66, 67, 93], "us": [0, 1, 2, 3, 4, 5, 6, 8, 9, 11, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 36, 40, 41, 45, 48, 55, 56, 57, 58, 64, 65, 66, 67, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 82, 83, 84, 85, 87, 88, 90, 92, 93, 94, 95, 96, 97, 98, 99], "usabl": 94, "usag": [0, 5, 7, 8, 16, 19, 21, 24, 28, 29, 30, 40, 64, 70, 73, 80, 81, 82, 88, 93, 97], "use_beam_hyp": 87, "use_beam_search": [49, 70, 93], "use_cach": [82, 83, 84], "use_context_fmha_for_gener": 93, "use_cuda_graph": [20, 27, 56, 70, 74], "use_custom_all_reduc": 93, "use_diff_of_squar": 82, "use_dynamic_tre": [43, 44, 70], "use_embedding_shar": 93, "use_fp32_acc": 82, "use_fp8": 83, "use_fp8_context_fmha": [5, 29, 73, 93], "use_fused_mlp": [29, 73, 93], "use_gemm_allreduce_plugin": 87, "use_gpt_attention_plugin": 87, "use_gpu_direct_storag": 87, "use_implicit_relative_attent": 83, "use_kv_cach": [70, 83, 87], "use_logn_sc": 83, "use_lora": 84, "use_lora_plugin": 87, "use_mamba_conv1d_plugin": 87, "use_meta_recip": 70, "use_modelopt_ckpt": 54, "use_modelopt_quant": 19, "use_mrop": 70, "use_one_more_block": 87, "use_paged_context_fmha": [5, 9, 29, 73, 77, 80], "use_parallel_embed": [15, 16, 84], "use_preload": 84, "use_prompt_tun": [84, 93], "use_py_sess": 92, "use_refit": 70, "use_relaxed_acceptance_for_think": [26, 27, 70], "use_runtime_default": 87, "use_safetensors_load": 84, "use_strip_plan": 70, "use_tqdm": 70, "use_variable_beam_width_search": 87, "usebantoken": 0, "usebanword": 0, "usecrossattent": 1, "usedefaultvalu": 1, "usednumblock": 0, "usedraftlogit": 1, "usedraftlogitshost": 1, "usedynamictre": 0, "usedynamictreehost": 1, "useexpliciteosstop": 0, "usefrequencypenalti": 0, "usegemmallreduceplugin": 1, "usegptattentionplugin": [1, 6], "usegpudirectstorag": 0, "uselanguageadapt": 1, "useloraplugin": 1, "usemambaconv1dplugin": 1, "usemaxlengthstop": 0, "useminlen": 0, "useminlength": 0, "useminp": 0, "usemrop": 1, "usenorepeatngrams": 0, "useoccurrencepenalti": 0, "usepackedinput": 1, "usepagedst": 1, "usepenalti": 0, "usepositionembed": 1, "usepresencepenalti": 0, "useprogthread": 0, "useprompttun": 1, "user": [0, 2, 3, 5, 6, 7, 9, 10, 11, 16, 17, 18, 19, 20, 24, 25, 26, 27, 28, 30, 31, 33, 34, 43, 44, 45, 49, 52, 53, 54, 60, 61, 65, 69, 70, 72, 73, 74, 79, 80, 81, 82, 84, 88, 89, 90, 92, 93], "user_buff": [29, 77], "userandomacceptancethreshold": 1, "userbuff": [70, 93], "userepetitionpenalti": 0, "userwarn": 67, "useshapeinfer": 1, "usespecdecod": 1, "usestopword": 0, "usetemp": 0, "usetemperatur": 0, "usetokentypeembed": 1, "usevariablebeamwidthsearch": 0, "usr": [15, 20, 30, 33, 34, 35, 37, 38, 39, 67, 73], "usual": [16, 19, 27, 67, 70, 74, 75, 80, 82, 98], "util": [0, 1, 2, 5, 6, 12, 16, 20, 21, 26, 28, 29, 40, 67, 71, 72, 73, 77, 80, 81, 89, 93, 97], "uv": 28, "uv_gemm": 26, "uvm": [0, 1], "v": [1, 2, 5, 6, 10, 20, 21, 22, 25, 26, 28, 64, 82, 84, 87, 90, 91, 92, 95, 97], "v0": [10, 21, 22, 23, 24, 71, 73, 74, 91, 93], "v1": [30, 33, 34, 35, 38, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 58, 59, 60, 61, 62, 66, 67, 69, 88, 91, 93, 94], "v10": 93, "v100": 93, "v12": 93, "v2": [25, 28, 90, 93], "v3": [27, 30, 72, 90, 91, 93], "v9": 23, "v_dim": 82, "v_head_dim": [82, 83], "v_proj": [17, 95], "vacat": [42, 46, 47, 49, 50], "valid": [0, 1, 3, 12, 27, 70, 74, 82, 87], "validate_cuda_graph_config": 70, "validate_cuda_graph_max_batch_s": 70, "validate_positive_valu": 70, "validatevec": 1, "validationerror": 70, "validmpiconfig": 1, "valu": [0, 1, 2, 5, 6, 8, 9, 10, 11, 13, 15, 16, 17, 20, 21, 22, 27, 28, 29, 30, 36, 59, 70, 73, 75, 77, 79, 81, 82, 84, 85, 86, 87, 89, 90, 92, 93, 97, 98, 99], "valuabl": 26, "value_typ": 0, "valuestatu": 1, "vanilla": [5, 97], "vanillaattent": 97, "var": 82, "vari": [23, 79, 80, 98], "variabl": [0, 1, 6, 8, 17, 20, 23, 26, 55, 56, 57, 64, 67, 70, 72, 73, 92, 93], "variabledraftlength": 1, "varianc": [28, 77, 79, 80, 82], "variant": [0, 3, 5, 19, 21, 27, 28, 69, 82, 88, 93, 97], "varieti": [73, 75, 93], "variou": [5, 12, 18, 73, 77, 79, 93], "varnam": 1, "vartyp": 1, "vboost": [20, 26, 73], "vbw": 93, "ve": [26, 58], "vec": [0, 1], "vec2": 82, "veclogprob": 0, "vectoken": 0, "vectokenextraid": [0, 1], "vector": [0, 1, 3, 5, 6, 8, 10, 28, 82], "vecuniquetoken": [0, 1], "verbatim": 84, "verbos": [29, 30, 73], "veri": [5, 15, 16, 18, 25, 27, 75, 76, 77, 93], "verif": [0, 12, 27, 70], "verifi": [12, 27, 64, 80, 82, 93], "verificationsets": 0, "versa": [9, 28], "version": [0, 1, 2, 5, 6, 15, 17, 19, 20, 26, 28, 30, 36, 65, 67, 73, 75, 82, 88, 92, 93, 94], "vertic": 82, "vertical_strid": 83, "vgqa": 8, "via": [0, 2, 11, 12, 26, 55, 56, 57, 58, 65, 67, 73, 77, 78, 80, 81, 82, 93, 94], "vice": [9, 28, 59], "vicuna": [12, 43, 44, 54], "video": [34, 61, 73, 87, 91, 93], "video_grid_thw": 87, "video_path": 87, "video_preprocess": 87, "video_url": [34, 61], "view": [1, 27, 82, 87], "vila": [34, 61, 90, 91, 93], "vinyl": 73, "violat": 93, "virtual": [0, 1, 83], "vision": [87, 90, 91, 93], "vision_grid_thw": 87, "vision_length": 82, "vision_model_typ": 84, "vision_start": 82, "vision_token_mask": 83, "visit": [12, 26, 93], "visual": [79, 93], "visual_engine_dir": 87, "visual_featur": 87, "visualize_network": [29, 70, 93], "vit": 93, "vital": [7, 25], "vl": [30, 34, 39, 61, 73, 91, 93], "vlm": [91, 93], "vocab": [82, 87], "vocab_embed": [14, 17], "vocab_s": [0, 15, 17, 70, 83, 84, 87, 95], "vocab_size_pad": 87, "vocabs": [1, 6], "vocabsizepad": [0, 1], "vocabulari": [0, 1, 6, 9, 12, 74, 83, 87], "void": [0, 1, 3, 16], "volta": 93, "volum": [1, 11, 65, 73], "volumenonneg": 1, "vonjackustc": 93, "vote": [42, 46, 47, 49, 50], "vswa": 8, "vulner": 93, "vultureprim": 93, "w": [1, 24, 26, 28, 30, 82, 84, 90, 91, 93], "w1": 82, "w4a": [90, 93], "w4a16": [15, 25, 64, 70, 84], "w4a16_awq": [15, 19, 36, 59, 70], "w4a16_gptq": [15, 70], "w4a8": [25, 93], "w4a8_awq": [15, 19, 70], "w4a8_qserve_per_channel": 70, "w4a8_qserve_per_group": 70, "w4aint8": 93, "w8a": 90, "w8a16": [15, 25, 64, 70, 84], "w8a16_gptq": 70, "w8a8": [22, 25, 64], "w8a8_sq_per_channel": [15, 70], "w8a8_sq_per_channel_per_tensor_plugin": [70, 84], "w8a8_sq_per_channel_per_token_plugin": [70, 84], "w8a8_sq_per_tensor_per_token_plugin": [70, 84], "w8a8_sq_per_tensor_plugin": [70, 84], "wa": [0, 1, 3, 5, 6, 15, 27, 28, 67, 69, 73, 74, 75, 77, 79, 80, 81, 83, 90, 92, 93, 95, 99], "wai": [2, 5, 7, 11, 18, 26, 27, 28, 50, 52, 69, 71, 73, 75, 77, 82, 89, 93], "wait": [0, 1, 3, 19, 28, 36, 70, 71, 73, 82], "walk": [34, 58, 61, 75, 76, 77], "wang1120": 93, "wangkuiyi": 93, "want": [5, 12, 19, 26, 27, 32, 67, 70, 72, 73, 77, 79, 81, 82, 92, 93, 95], "warm": 98, "warmup": [20, 72, 73, 75, 93, 97, 98], "warn": [5, 29, 30, 70, 73, 74, 89], "warp": [11, 93], "wast": 28, "watch": 80, "wdkv": 26, "wdq": 26, "we": [1, 2, 4, 6, 7, 10, 11, 12, 13, 15, 19, 20, 24, 25, 26, 27, 28, 30, 31, 32, 42, 46, 47, 49, 50, 58, 59, 65, 67, 69, 72, 73, 74, 75, 76, 77, 79, 80, 82, 87, 88, 92, 93, 95], "weapon": 51, "wear": 51, "web": [18, 32], "weig": 82, "weight": [0, 1, 4, 10, 19, 21, 22, 25, 26, 27, 29, 30, 50, 64, 70, 71, 74, 75, 76, 77, 82, 83, 84, 87, 88, 93], "weight_index": 82, "weight_load": 83, "weight_only_groupwise_quant_matmul": 90, "weight_only_precis": 93, "weight_spars": [29, 70], "weight_stream": [13, 29, 70], "weightonlygroupwisequantmatmulplugin": 90, "weights_dict": 19, "weights_scaling_factor": [15, 17], "weightsinpoint": 1, "weightsoutpoint": 1, "well": [5, 6, 16, 18, 22, 36, 72, 79, 80, 90, 91], "were": [0, 1, 12, 15, 19, 21, 25, 28, 74, 76, 79, 93], "weren": 67, "wget": 92, "what": [2, 3, 28, 34, 58, 61, 64, 65, 72, 73, 75, 77, 79, 80], "whatev": 1, "wheel": [65, 67, 93], "when": [0, 1, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 16, 17, 19, 20, 24, 25, 27, 28, 29, 31, 36, 52, 64, 65, 67, 70, 72, 73, 75, 77, 79, 80, 81, 82, 83, 84, 87, 88, 89, 90, 92, 93, 95, 97, 98], "whenev": 1, "where": [0, 1, 2, 5, 6, 8, 9, 11, 12, 15, 16, 21, 25, 26, 27, 28, 30, 33, 35, 36, 58, 60, 62, 70, 73, 74, 77, 79, 81, 82, 87, 88, 90, 93, 99], "wherea": [0, 15, 79], "whether": [0, 1, 2, 3, 5, 10, 29, 70, 76, 77, 80, 82, 83, 87, 96, 97], "which": [0, 1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 15, 16, 17, 19, 21, 25, 26, 27, 28, 29, 30, 65, 67, 69, 70, 72, 73, 75, 77, 79, 80, 81, 82, 84, 85, 87, 88, 89, 90, 93, 94, 96, 97, 99], "while": [0, 1, 4, 7, 8, 9, 11, 12, 16, 19, 21, 22, 24, 25, 26, 27, 28, 67, 71, 73, 75, 76, 77, 78, 79, 80, 81, 82, 89, 90, 93, 97], "whisper": [90, 91, 93], "whisperencod": 84, "whl": [20, 65, 66, 67], "who": [27, 69], "whole": [1, 70, 71, 82], "whose": [2, 9, 15, 26, 83], "why": [0, 2, 16, 28, 70, 77, 79, 80, 82, 89], "wide": [0, 4, 27, 70, 75], "width": [0, 1, 5, 6, 39, 70, 83, 87, 89, 93], "win": 70, "window": [0, 1, 8, 12, 29, 64, 70, 73, 82, 87, 93], "window_s": 5, "windows": 0, "wip": 26, "wireless": 45, "wirelessaccesspoint": 45, "wise": [7, 70, 82, 93], "wish": 9, "wit": 51, "with_ssh": 31, "within": [1, 2, 5, 8, 11, 12, 16, 28, 51, 70, 73, 76, 77, 79, 80, 82, 88, 98], "without": [0, 1, 3, 5, 11, 12, 16, 17, 20, 25, 26, 29, 36, 51, 71, 73, 77, 80, 82, 84, 93, 95, 97], "wkr": 26, "wo": [17, 26, 93], "wo_gemm": 26, "won": [67, 76], "word": [0, 3, 5, 70, 82, 87, 93], "word_dict": 87, "word_embed": 17, "word_embeddings_layernorm": 17, "work": [5, 6, 7, 8, 11, 12, 16, 19, 20, 36, 52, 55, 56, 57, 59, 65, 67, 71, 74, 78, 82, 87, 90, 92, 93, 95], "workaround": [17, 20, 93], "workdir": [30, 55, 56, 57, 65], "worker": [16, 29, 30, 70, 73, 89, 93], "workerexecutablepath": 0, "workflow": [5, 6, 14, 15, 20, 27, 36, 64, 69, 74, 75, 77, 78, 82, 88, 92, 93, 94], "workload": [4, 11, 16, 28, 29, 72, 73, 75, 77, 78, 79, 80], "workspac": [1, 29, 30, 70, 73, 82, 89, 93], "workstat": 22, "world": [0, 2, 7, 20, 27, 29, 55, 56, 57, 71, 73, 75, 76, 77, 82], "world_config": 87, "world_siz": [15, 19, 82, 93], "worldconfig": [0, 6, 87], "worldsiz": 1, "wors": [12, 29, 77], "worst": [79, 80], "worth": [5, 8, 77, 80], "would": [0, 7, 12, 27, 73, 75, 77, 79, 81, 82, 95], "wouldn": 51, "wpa2": 45, "wqr": 26, "wrap": [0, 1, 16, 29, 69, 75, 82, 85, 87, 93], "wrapped_properti": 70, "wrapper": [1, 7, 19, 97], "write": [0, 1, 9, 17, 26, 29, 64, 82, 92], "written": [16, 73, 82], "wrong": [12, 51, 93], "wsl": 93, "wuk": 26, "wuq": 26, "wuv": 26, "www": 93, "x": [0, 1, 3, 6, 10, 13, 30, 73, 82, 83, 84, 88, 90, 93], "x86": 9, "x86_64": 91, "xcomposer2": 93, "xgrammar": [0, 3, 45, 93], "xl": 93, "xml": 3, "xor": 82, "xqa": 93, "xxx": [17, 19, 92], "xxx_plugin": 85, "xy": 82, "y": [2, 3, 20, 24, 31, 65, 66, 67, 73, 82, 84, 90], "y_bia": 82, "yaml": [30, 73, 74], "yarn": 82, "ye": [2, 82, 89], "yeah": 58, "yelp": 91, "yen": 73, "yet": [0, 6, 19, 22, 26, 82, 99], "yield": [9, 28, 36, 77, 79], "yiyixu": [34, 61], "yml": [20, 27, 30, 37, 73, 74], "york": [30, 33, 35, 60, 62, 88], "you": [3, 4, 5, 6, 7, 9, 10, 12, 15, 16, 18, 19, 20, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 36, 45, 46, 49, 52, 55, 56, 57, 58, 59, 60, 61, 64, 65, 67, 69, 70, 73, 74, 76, 77, 78, 79, 80, 81, 82, 87, 88, 89, 92, 93, 94, 95, 97], "your": [9, 10, 11, 12, 18, 19, 20, 25, 27, 29, 31, 32, 36, 58, 65, 67, 69, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 88, 92, 95, 97, 98], "your_data_path": [20, 27], "your_dockerhub_usernam": [31, 32], "your_model_dir": 27, "your_model_path": 20, "your_public_kei": 32, "your_work_path": 20, "yourself": 94, "yuhuili": [43, 44], "yyi": 92, "z": 82, "zars19": 93, "zero": [0, 1, 3, 17, 69, 70, 82, 83, 90, 92], "zero_is_placehold": 82, "zip": 52, "zjli2013": 93, "zoo": 93}, "titles": ["Executor", "Runtime", "Disaggregated-Service (experimental)", "Executor API", "Expert Parallelism in TensorRT-LLM", "Multi-Head, Multi-Query, and Group-Query Attention", "C++ GPT Runtime", "Graph Rewriting Module", "KV Cache Management: Pools, Blocks, and Events", "KV cache reuse", "Run gpt-2b + LoRA using Executor / cpp runtime", "Low-Precision-AllReduce", "Speculative Sampling", "Running With Weight Streaming to Reduce GPU Memory Consumption", "Adding a Model", "TensorRT-LLM Checkpoint", "Model Definition", "TensorRT-LLM Model Weights Loader", "TensorRT-LLM Architecture", "TensorRT-LLM Build Workflow", "How to get best performance on DeepSeek-R1 in TensorRT-LLM", "Falcon-180B on a single H200 GPU with INT4 AWQ, and 6.7x faster Llama-70B over A100", "H100 has 4.6x A100 Performance in TensorRT-LLM, achieving 10,000 tok/s at 100ms to first token", "H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLM", "New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget", "Speed up inference with SOTA quantization techniques in TRT-LLM", "Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs", "DeepSeek R1 MTP Implementation and Optimization", "Optimizing DeepSeek R1 Throughput on NVIDIA Blackwell GPUs: A Deep Dive for Developers", "trtllm-build", "trtllm-serve", "Build the TensorRT-LLM Docker Image", "Develop TensorRT-LLM on Runpod", "Curl Chat Client", "Curl Chat Client For Multimodal", "Curl Completion Client", "LLM Common Customizations", "Deepseek R1 Reasoning Parser", "Genai Perf Client", "Genai Perf Client For Multimodal", "LLM Examples Introduction", "LLM Examples", "Automatic Parallelism with LLM", "Generate Text Using Eagle2 Decoding", "Generate Text Using Eagle Decoding", "Generate text with guided decoding", "Generate text", "Generate Text Asynchronously", "Generate Text in Streaming", "Generate text with customization", "Distributed LLM Generation", "Get KV Cache Events", "Control generated text using logits processor", "Generate Text Using Lookahead Decoding", "Generate Text Using Medusa Decoding", "Llm Mgmn Llm Distributed", "Llm Mgmn Trtllm Bench", "Llm Mgmn Trtllm Serve", "Generate text with multiple LoRA adapters", "Generation with Quantization", "OpenAI Chat Client", "OpenAI Chat Client", "OpenAI Completion Client", "Online Serving Examples", "Welcome to TensorRT-LLM\u2019s Documentation!", "Building from Source Code on Linux", "Installing on Grace Hopper", "Installing on Linux", "Key Features", "API Introduction", "API Reference", "Overview", "Performance Analysis", "TensorRT-LLM Benchmarking", "Overview", "Benchmarking Default Performance", "Deciding Model Sharding Strategy", "FP8 Quantization", "Performance Tuning Guide", "Tuning Max Batch Size and Max Num Tokens", "Useful Build-Time Flags", "Useful Runtime Options", "Functionals", "Layers", "Models", "Plugin", "Quantization", "Runtime", "Quick Start Guide", "Memory Usage of TensorRT-LLM", "Numerical Precision", "Support Matrix", "Troubleshooting", "Release Notes", "PyTorch Backend", "Adding a New Model in PyTorch Backend", "Architecture Ovewiew", "Attention", "KV Cache Manager", "Scheduler"], "titleterms": {"": [5, 22, 25, 64], "0": 93, "000": [22, 23], "1": [14, 16, 20, 65, 74, 89, 93], "10": [22, 93], "100m": 22, "11": 93, "12": [23, 93], "13": 93, "13b": 23, "14": 93, "15": 93, "16": 93, "17": 93, "18": 93, "180b": 21, "19": 93, "2": [14, 20, 24, 65, 89, 93], "2b": 10, "3": [14, 16, 20, 73, 74, 89, 91], "4": [14, 20, 22], "405b": [16, 74], "4x": 24, "5": 20, "6": [20, 21], "6x": 22, "7": 93, "70b": [16, 21, 24, 73, 74], "7x": 21, "8": 93, "8b": 74, "9": 93, "A": 28, "As": 3, "For": [34, 39], "In": [3, 5, 71], "Not": 89, "One": [26, 65], "The": [3, 90], "To": 75, "With": [13, 71], "a100": [21, 22], "about": [12, 30, 71, 76], "absorb": 28, "accept": [26, 27], "access": 31, "account": 32, "accuraci": [11, 25, 27], "achiev": [22, 23, 27], "acknowledg": [26, 27, 28], "activ": [83, 89], "ad": [14, 95], "adapt": 58, "addit": 3, "adp": 28, "advanc": 64, "algorithm": 11, "alibi": 5, "allreduc": 11, "an": 8, "analysi": 72, "announc": 93, "api": [3, 7, 13, 19, 30, 40, 69, 70, 75, 88, 93, 96], "arbitrari": 3, "architectur": [18, 26, 64, 96], "argument": 29, "asynchron": 47, "asyncio": 36, "attent": [5, 15, 26, 27, 28, 71, 79, 80, 81, 83, 97], "attentionbackend": 97, "attentionmetadata": 97, "auto": 29, "automat": 42, "autoregress": 26, "avoid": 75, "awq": [15, 21, 90], "b200": [20, 26], "backend": [26, 91, 94, 95, 97], "background": [26, 27], "balanc": 26, "base": [27, 36], "baselin": 77, "basic": 27, "batch": [3, 5, 71, 79], "beam": [3, 5], "befor": [73, 75], "begin": 75, "behavior": 73, "bench": [56, 72, 75], "benchmark": [2, 20, 25, 30, 73, 74, 75], "best": [20, 25], "bf16": 90, "bia": 5, "bind": [3, 16, 65], "blackwel": [28, 90], "block": 8, "blockmanag": 8, "boost": 73, "boundari": 26, "budget": 24, "buffer": [5, 77, 89], "buffermanag": 1, "build": [15, 19, 20, 29, 31, 32, 36, 65, 73, 75, 80], "c": [3, 6, 65, 89], "cach": [5, 8, 9, 15, 51, 77, 81, 89, 98], "cachecommun": 0, "can": [9, 71], "capac": 81, "case": 79, "cast": 83, "caveat": 73, "chang": [13, 79, 93], "chat": [30, 33, 34, 60, 61], "checkpoint": 15, "choos": 25, "chunk": [5, 20, 79, 81], "class": 3, "classic": 7, "cli": [19, 75], "client": [33, 34, 35, 38, 39, 60, 61, 62], "clock": [20, 73], "close": [21, 24], "code": 65, "collect": 72, "combin": 20, "come": 25, "command": 74, "common": [1, 36, 71], "commun": [26, 76], "compil": [16, 20, 65, 88], "complet": [30, 35, 62], "compon": [6, 94], "conclus": [77, 79, 80], "config": [15, 29], "configur": [3, 6, 10, 26, 32, 36, 77, 80, 95], "connect": 32, "consider": 11, "consumpt": 13, "contain": [20, 31, 65], "content": [20, 26, 27, 28, 78, 95], "context": [3, 5, 20, 79, 80, 81], "contigu": 5, "control": [3, 52], "conv": 83, "convers": [14, 19], "coordin": 72, "core": 95, "cpp": 10, "creat": [32, 65], "cross": 5, "cuda": 26, "cudaev": 1, "cudastream": 1, "curl": [33, 34, 35], "custom": [17, 36, 49, 98, 99], "cutlass": 26, "cyclic": 5, "data": 28, "dataset": [20, 73, 74, 75], "datatransceiverst": 0, "debug": [2, 72, 92], "decid": 76, "decod": [3, 12, 27, 29, 43, 44, 45, 53, 54, 89, 96], "decoderst": 1, "decodinginput": 1, "decodingoutput": 1, "decor": 7, "deep": 28, "deepseek": [20, 26, 27, 28, 37], "default": [20, 26, 73, 75], "definit": [16, 88, 95], "dens": 26, "depend": 26, "deploi": 88, "dequant": 90, "descript": 72, "detail": [10, 90], "develop": [28, 32, 94], "diagram": 26, "differ": 3, "disabl": 36, "disaggreg": [2, 30], "disaggregated_mpi_work": 30, "disaggserverutil": 0, "distribut": [50, 55], "dive": 28, "do": 71, "docker": [31, 32, 65], "dockerhub": [31, 32], "document": [64, 93], "dora": 10, "download": 20, "dq": 90, "draft": 12, "e2": 92, "eagl": [12, 27, 44], "eagle2": 43, "eagle3": 27, "eaglebuff": 1, "eaglemodul": 1, "embed": [5, 83], "enabl": [4, 9, 20, 31, 72, 77, 80], "endpoint": 30, "engin": [15, 16, 69, 73, 75, 88, 96], "enhanc": 93, "environ": [2, 11], "ep": 28, "error": 92, "etp": 26, "evalu": [15, 27], "event": [8, 51], "everyth": 26, "exampl": [2, 3, 10, 15, 16, 17, 40, 41, 63, 72, 73], "except": 89, "execut": 92, "executor": [0, 3, 10], "expect": [9, 20], "experiment": 2, "expert": [4, 26, 28], "explicitdrafttokensbuff": 1, "explor": 20, "face": 69, "factor": [5, 15], "falcon": 21, "faq": [2, 89], "faster": 21, "featur": [20, 68, 72, 93], "file": 65, "first": 22, "fix": [27, 93], "flag": [80, 90], "flayerinfo": 7, "flight": [3, 5, 71], "flow": 73, "fmha": 5, "format": [10, 20], "fp16": 90, "fp32": 90, "fp4": 74, "fp8": [5, 15, 22, 71, 74, 77, 90], "fraction": 81, "free": 81, "from": 65, "full": 65, "fulli": 17, "function": [7, 17, 82], "fuse_a_gemm": 26, "fusion": [16, 26, 77, 80], "futur": [26, 27, 28, 36], "garbag": 72, "gate": 77, "gc": 72, "gemm": [26, 77, 80], "genai": [38, 39], "gener": [2, 5, 36, 43, 44, 45, 46, 47, 48, 49, 50, 52, 53, 54, 58, 59], "get": [20, 51, 64], "gil": 72, "gpt": [6, 10], "gptdecod": 1, "gptdecoderbatch": 1, "gptjsonconfig": 1, "gptq": 90, "gpu": [13, 16, 20, 21, 26, 28, 71, 73, 81, 89], "grace": 66, "graph": [7, 26], "group": [5, 26], "guid": [3, 45, 78, 88, 94, 95], "h": [0, 1], "h100": [22, 23], "h200": [20, 21, 23, 24], "ha": 22, "hardwar": 91, "hbm": 23, "head": 5, "header": 65, "hierarchi": 8, "high": 7, "hopper": [66, 90], "host": 9, "how": [4, 9, 20, 26, 27, 28, 73, 76, 79], "hub": 69, "hug": 69, "i": [22, 76, 89], "ibuff": 1, "id": 10, "igptdecoderbatch": 1, "imag": [31, 32, 65], "implement": [14, 26, 27, 97], "import": 5, "improv": 12, "increas": 24, "indic": 64, "infer": [3, 25, 27, 30, 71, 88, 89], "inform": [7, 72, 88], "infrastructur": 93, "input": 5, "instal": [20, 64, 66, 67, 92], "int4": [21, 90], "int8": [5, 90], "interfac": 98, "intern": 6, "introduct": [28, 40, 69, 95, 98, 99], "ipcnvlsmemori": 1, "ipcutil": 1, "isl": 20, "issu": [20, 27, 89, 93, 94], "itensor": 1, "iter": 72, "kei": [17, 26, 32, 68, 76, 93, 94], "kernel": [24, 26], "knowledg": 78, "known": [27, 65, 89, 93, 94], "kv": [5, 8, 9, 15, 51, 77, 81, 89, 98], "kvcacheeventmanag": 8, "kvcachemanag": 96, "latenc": [20, 24, 26, 73, 75, 77], "latest": [23, 71], "launch": [26, 72], "layer": [26, 28, 83], "layernorm": 15, "layout": 17, "level": [7, 26, 96], "limit": [12, 65, 73, 93], "linear": 83, "link": 65, "linux": [65, 67], "llama": [16, 21, 24, 73, 74, 77, 80], "llama2": 23, "llm": [4, 12, 15, 17, 18, 19, 20, 22, 23, 25, 27, 31, 32, 36, 40, 41, 42, 50, 55, 56, 57, 64, 65, 69, 71, 73, 75, 79, 88, 89, 91, 93], "load": [17, 95], "loader": 17, "local": 69, "logit": [3, 29, 52], "lookahead": [12, 53], "lookaheadbuff": 1, "lookaheadmodul": 1, "lookup": 12, "lora": [10, 29, 58], "loracach": [1, 10], "loracachepagemanagerconfig": 1, "loramodul": 1, "low": [11, 73, 77], "make": 15, "manag": [7, 8, 73, 98], "map": [10, 73], "mark": 3, "marker": 72, "match": 16, "matrix": [90, 91], "max": [20, 73, 79, 81], "maximum": 81, "measur": 74, "medusa": [12, 54, 73], "medusamodul": 1, "memori": [9, 13, 20, 23, 81, 89], "memorycount": 1, "method": [7, 25], "metric": 30, "mgmn": [55, 56, 57], "min": 20, "mix": 26, "mixtur": 4, "mla": 28, "mlp": [15, 77, 83], "mlperf": 22, "modal": [73, 91], "mode": 73, "model": [6, 12, 14, 16, 17, 18, 20, 26, 27, 69, 73, 74, 76, 77, 80, 84, 88, 91, 92, 93, 95, 96], "modelconfig": 1, "modul": [7, 10, 27, 28], "moe": [4, 28], "moe_backend": 26, "more": [20, 24, 72], "mqa": 28, "mtp": [26, 27], "multi": [5, 16, 26, 30, 71, 73, 91], "multimod": [30, 34, 39], "multipl": [58, 80], "name": [17, 29], "nativ": [17, 71], "nearli": 23, "network": 73, "new": [14, 24, 95, 97], "next": [25, 88], "node": [16, 30, 71], "non": 73, "norm": [77, 80], "normal": 83, "note": [3, 5, 93], "nsight": 72, "num": 79, "numer": 90, "nvfp4": 90, "nvidia": [26, 28, 72], "nvtx": 72, "o": 89, "obtain": 3, "offload": 9, "onli": [26, 65, 72, 90], "onlin": 63, "openai": [60, 61, 62], "optim": [5, 26, 27, 28, 80], "option": [20, 65, 77, 80, 81], "osl": 20, "other": 73, "out": [20, 95], "output": [3, 73], "over": 21, "overview": [6, 15, 17, 19, 71, 74], "ovewiew": 96, "own": 99, "p": 9, "pack": 5, "pad": 5, "page": [5, 8, 71, 79, 80, 81], "parallel": [4, 10, 26, 28, 29, 42, 73, 76, 80], "paramet": 6, "parser": 37, "part": 14, "pattern": [7, 16], "perf": [38, 39], "perform": [9, 11, 12, 20, 22, 25, 26, 64, 72, 75, 77, 78, 80], "persist": 73, "phase": 5, "pipelin": [76, 80], "pitfal": 75, "plugin": [16, 29, 77, 80, 85], "pod": 32, "polici": 81, "pool": [8, 83, 89], "posit": 5, "post": 3, "postprocess": 17, "power": 73, "practic": 25, "precis": [11, 26, 28, 90], "prepar": [15, 20, 32, 69, 73, 74, 75], "prerequisit": [20, 65, 78, 88, 95], "prevent": 9, "processor": [3, 52], "profil": [26, 72, 80], "programmat": 26, "prompt": 12, "prompttuningparam": 1, "provid": 24, "push": 26, "pyexecutor": 96, "python": [3, 65, 89], "pytorch": [72, 73, 91, 94, 95], "q": 90, "qkv": 5, "quantiz": [15, 19, 25, 36, 59, 73, 77, 86, 90, 94], "quantmod": 90, "queri": 5, "quick": [88, 94], "quickstart": 73, "r1": [20, 26, 27, 28, 37], "rab": 5, "rank": 15, "rawengin": 1, "re": 26, "reason": 37, "recommend": [77, 80, 89], "record_signatur": 7, "redraft": 12, "reduc": [13, 77, 80], "refer": [14, 64, 70], "regist": 14, "registr": 95, "rel": 5, "relat": [7, 88], "relax": [26, 27], "releas": 93, "reproduc": [20, 26, 28, 74], "request": [1, 3], "requir": [7, 11], "resourcemanag": 96, "respons": 3, "result": [3, 20, 72, 74, 75], "retriev": 7, "reus": 9, "revisit": 79, "rewrit": 7, "right": 25, "roll": 5, "rope": 5, "rotari": 5, "router": 26, "routergemm": 26, "run": [10, 13, 20, 27, 72, 73, 74, 75, 88], "runpod": 32, "runtim": [1, 6, 10, 16, 28, 36, 65, 81, 87, 89], "runtimedefault": 1, "same": 24, "sampl": [6, 12, 36], "samplingconfig": 1, "save": 75, "scale": [5, 15], "scatter": 80, "schedul": [79, 81, 96, 99], "script": [41, 63], "search": 5, "sec": 23, "send": 3, "serial": 0, "serv": [30, 57, 63, 72, 88], "server": [3, 30, 88], "servic": 2, "set": [73, 76], "shard": 76, "shoot": 17, "singl": 21, "situat": 9, "size": [79, 81, 89], "slide": 5, "slurm": 30, "smart": 26, "smoothquant": 90, "softwar": 91, "sota": 25, "sourc": 65, "spars": 26, "specif": 72, "specul": [12, 27, 29], "speculativedecodingmod": 1, "speculativedecodingmodul": 1, "speed": 25, "speedup": 27, "ssh": [31, 32], "start": [30, 64, 88, 94], "step": [14, 20, 65, 88, 95], "strategi": [26, 28, 76], "stream": [13, 26, 48], "streamingllm": 5, "structur": 3, "studi": [27, 79], "style": 36, "subcommand": 73, "summari": [73, 77, 80], "support": [16, 17, 20, 27, 65, 69, 71, 73, 90, 91], "swiglu": 77, "syntax": 30, "system": [26, 72], "tabl": [20, 26, 27, 28, 64, 78, 95], "target": 12, "technic": 90, "techniqu": 25, "templat": 32, "tensor": [0, 3, 4, 5, 7, 10, 76, 89], "tensorrt": [4, 12, 15, 16, 17, 18, 19, 20, 22, 23, 25, 27, 31, 32, 64, 65, 69, 71, 73, 75, 79, 88, 89, 91, 93], "test": 92, "text": [43, 44, 45, 46, 47, 48, 49, 52, 53, 54, 58], "think": 76, "throughput": [20, 24, 28, 73, 74, 75], "time": [80, 89], "tip": [69, 75, 92], "tllmlogger": 1, "tok": 22, "token": [22, 23, 36, 79, 81], "tool": 19, "top": 96, "topologi": 11, "transferag": 0, "translat": 17, "tree": [12, 27, 95], "trigger": 8, "triton": [3, 88], "troubl": 17, "troubleshoot": [2, 69, 75, 92], "trt": 25, "trtllm": [26, 29, 30, 56, 57, 72, 75, 88], "tune": [9, 20, 78, 79], "type": [0, 8], "understand": [79, 89], "unit": 92, "up": [21, 24, 25], "updat": 93, "upload": [31, 32], "us": [7, 10, 12, 43, 44, 52, 53, 54, 80, 81, 89], "usag": [2, 11, 89], "user": 77, "v": [4, 23], "valid": 73, "vanilla": 27, "variabl": [2, 11, 74], "verif": 26, "verifi": 14, "via": 75, "visual": 72, "w4a16": 90, "w8a16": 90, "w8a8": 90, "weight": [13, 14, 15, 16, 17, 18, 28, 89, 90, 95], "welcom": 64, "what": [8, 22, 25, 71], "when": [7, 26], "width": 3, "window": [5, 71, 81], "windowblockmanag": 8, "wip": 20, "within": 24, "without": 65, "work": [26, 27, 28, 73], "workflow": [7, 17, 19, 72, 73], "workload": 26, "world": 6, "worldconfig": 1, "write": 14, "xqa": [5, 24], "you": [71, 75], "your": 99}}) \ No newline at end of file diff --git a/latest/torch.html b/latest/torch.html index a67602b722..e23065ca67 100644 --- a/latest/torch.html +++ b/latest/torch.html @@ -51,7 +51,7 @@ @@ -63,7 +63,7 @@ - + @@ -336,6 +336,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -357,6 +358,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -421,6 +423,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -455,6 +458,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -730,6 +734,15 @@ scripts/huggingface_example.sh --model +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/torch/adding_new_model.html b/latest/torch/adding_new_model.html index b728120d1e..07ec6e7360 100644 --- a/latest/torch/adding_new_model.html +++ b/latest/torch/adding_new_model.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -355,6 +356,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -419,6 +421,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -453,6 +456,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -841,6 +845,15 @@

          +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/torch/arch_overview.html b/latest/torch/arch_overview.html index e6f16a6162..cbd02c40c2 100644 --- a/latest/torch/arch_overview.html +++ b/latest/torch/arch_overview.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -355,6 +356,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -419,6 +421,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -453,6 +456,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -707,6 +711,15 @@ The document +
          +
          + +

          Last updated on June 03, 2025.

          + +

          This page is generated by TensorRT-LLM commit 9ae2ce6.

          + +
          + diff --git a/latest/torch/attention.html b/latest/torch/attention.html index 6690997c5c..3f4be2531c 100644 --- a/latest/torch/attention.html +++ b/latest/torch/attention.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -355,6 +356,7 @@
        • Generate Text Asynchronously
        • Distributed LLM Generation
        • Control generated text using logits processor
          • +
        • Generate Text Using Eagle2 Decoding
        • Get KV Cache Events
        • Generate Text Using Lookahead Decoding
        • Generation with Quantization
          • @@ -419,6 +421,7 @@
        • Graph Rewriting Module
        • Run gpt-2b + LoRA using Executor / cpp runtime
        • Expert Parallelism in TensorRT-LLM
          • +
        • KV Cache Management: Pools, Blocks, and Events
        • KV cache reuse
        • Speculative Sampling
        • Disaggregated-Service (experimental)
          • @@ -453,6 +456,7 @@
        • Speed up inference with SOTA quantization techniques in TRT-LLM
        • New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
        • Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
          • +
        • DeepSeek R1 MTP Implementation and Optimization
          • @@ -518,7 +522,7 @@ The following sections explain how to use these implementations and provide a br

          Attention Backends#

          There are currently three available attention backends: the vanilla backend, the TRT-LLM backend, and the Flashinfer backend. -You can specify the desired attention backend using PyTorchConfig.attn_backend. For instance, to utilize the Flashinfer backend, you can create a PyTorchConfig with attn_backend = "flashinfer" and then pass it to the LLM constructor as follows: LLM(pytorch_backend_config=pytorch_config). This will enable the use of the Flashinfer backend for your model.

          +You can specify the desired attention backend using PyTorchConfig.attn_backend. For instance, to utilize the Flashinfer backend, you can pass attn_backend="flashinfer" to the LLM constructor as follows: LLM(attn_backend="flashinfer"). This will enable the use of the Flashinfer backend for your model.

          The vanilla backend, VanillaAttention, is a reference implementation designed primarily for inflight batching and linear KV cache support. While it serves as a useful baseline, it is not recommended for production use due to its limited optimizations.

          In contrast, the Flashinfer backend, FlashInferAttention, is performance-optimized and supports both inflight batching and paged KV cache. It also includes the following advanced features:

            @@ -831,6 +835,15 @@ For example, the Flashinfer metadata fills +
            + +

            Last updated on June 03, 2025.

            + +

            This page is generated by TensorRT-LLM commit 9ae2ce6.

            + +
            + diff --git a/latest/torch/kv_cache_manager.html b/latest/torch/kv_cache_manager.html index 0f379514aa..d37fa73257 100644 --- a/latest/torch/kv_cache_manager.html +++ b/latest/torch/kv_cache_manager.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
          1. Generate Text Asynchronously
          2. Distributed LLM Generation
          3. Control generated text using logits processor
            4. +
          4. Generate Text Using Eagle2 Decoding
          5. Get KV Cache Events
          6. Generate Text Using Lookahead Decoding
          7. Generation with Quantization
            8. @@ -355,6 +356,7 @@
          8. Generate Text Asynchronously
          9. Distributed LLM Generation
          10. Control generated text using logits processor
            11. +
          11. Generate Text Using Eagle2 Decoding
          12. Get KV Cache Events
          13. Generate Text Using Lookahead Decoding
          14. Generation with Quantization
            15. @@ -419,6 +421,7 @@
          15. Graph Rewriting Module
          16. Run gpt-2b + LoRA using Executor / cpp runtime
          17. Expert Parallelism in TensorRT-LLM
            18. +
          18. KV Cache Management: Pools, Blocks, and Events
          19. KV cache reuse
          20. Speculative Sampling
          21. Disaggregated-Service (experimental)
            22. @@ -453,6 +456,7 @@
          22. Speed up inference with SOTA quantization techniques in TRT-LLM
          23. New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
          24. Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
            25. +
          25. DeepSeek R1 MTP Implementation and Optimization
            26. @@ -509,6 +513,7 @@

            In Transformer-based models, the KV (Key-Value) Cache is a mechanism used to optimize decoding efficiency, particularly during autoregressive generation tasks. Since KV Cache requires memory to store, it is also an important resource. In TensorRT-LLM, KV Cache is managed by the KVCacheManager.

            +

            For details of the TensorRT-LLM KVCacheManager implementation see KV Cache Management.

            KV Cache Manager Introduction#

            KVCacheManager is a type of resource manager, inheriting from BaseResourceManager. @@ -699,6 +704,15 @@ Then, test it to ensure the +

            +
            + +

            Last updated on June 03, 2025.

            + +

            This page is generated by TensorRT-LLM commit 9ae2ce6.

            + +
            + diff --git a/latest/torch/scheduler.html b/latest/torch/scheduler.html index 912efd872e..225fef0c00 100644 --- a/latest/torch/scheduler.html +++ b/latest/torch/scheduler.html @@ -51,7 +51,7 @@ @@ -61,7 +61,7 @@ - + @@ -334,6 +334,7 @@
          26. Generate Text Asynchronously
          27. Distributed LLM Generation
          28. Control generated text using logits processor
            29. +
          29. Generate Text Using Eagle2 Decoding
          30. Get KV Cache Events
          31. Generate Text Using Lookahead Decoding
          32. Generation with Quantization
            33. @@ -355,6 +356,7 @@
          33. Generate Text Asynchronously
          34. Distributed LLM Generation
          35. Control generated text using logits processor
            36. +
          36. Generate Text Using Eagle2 Decoding
          37. Get KV Cache Events
          38. Generate Text Using Lookahead Decoding
          39. Generation with Quantization
            40. @@ -419,6 +421,7 @@
          40. Graph Rewriting Module
          41. Run gpt-2b + LoRA using Executor / cpp runtime
          42. Expert Parallelism in TensorRT-LLM
            43. +
          43. KV Cache Management: Pools, Blocks, and Events
          44. KV cache reuse
          45. Speculative Sampling
          46. Disaggregated-Service (experimental)
            47. @@ -453,6 +456,7 @@
          47. Speed up inference with SOTA quantization techniques in TRT-LLM
          48. New XQA-kernel provides 2.4x more Llama-70B throughput within the same latency budget
          49. Pushing Latency Boundaries: Optimizing DeepSeek-R1 Performance on NVIDIA B200 GPUs
            50. +
          50. DeepSeek R1 MTP Implementation and Optimization
            51. @@ -725,6 +729,15 @@ In the create_pytor

            +
            +
            + +

            Last updated on June 03, 2025.

            + +

            This page is generated by TensorRT-LLM commit 9ae2ce6.

            + +
            +
          -
        • visual_engine_dir (tensorrt_llm.runtime.MultimodalModelRunner property) -
          		•