Update latest GitHub pages to v1.2.0rc3

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: a9c5c8c57021602368f541d74d22523d
+config: 5b10b2153627779ea5be4dbb07d82396
 tags: 645f666f9bcd5a90fca523b33c5a78b7

latest/_downloads/b509390ba70e52fabb10dbd9d15d5118/attention.py

@@ -23,7 +23,7 @@ from ..distributed import AllReduceParams, alltoall_helix
 from ..model_config import ModelConfig
 from ..peft.lora.layer import LoraLayer, LoraModuleType
 from ..utils import (Fp4QuantizedTensor, get_model_extra_attrs,
-                     is_piecewise_running, is_torch_compiling)
+                     is_torch_compiling, maybe_compile)
 from .linear import Linear, TensorParallelMode, WeightMode, WeightsLoadingConfig
 from .multi_stream_utils import maybe_execute_in_parallel
 from .rms_norm import RMSNorm
@@ -76,17 +76,6 @@ def extract_extra_attrs(layer_idx: str, attn_type: str):
     return metadata, attn_layer
 
 
-def maybe_compile(func):
-
-    def wrapper(*args, **kwargs):
-        if is_piecewise_running():
-            # When piecewise running, we don't need to compile the function to avoid host overhead in attention op.
-            return func(*args, **kwargs)
-        return torch.compile(func)(*args, **kwargs)
-
-    return wrapper
-
-
 @maybe_compile
 def maybe_compiled_copy_(dst, src):
     dst.copy_(src)
@@ -413,9 +402,9 @@ class Attention(nn.Module):
 
         out_scale = None
         out_scale_sf = None
-        if self.has_quant_scale:
+        if self.has_quant_scale and not self.attn_output_gate:
             out_scale = self.o_proj.inv_input_scale
-        if self.o_proj.has_nvfp4 and self.support_nvfp4_output and enable_attn_nvfp4_output:
+        if self.o_proj.has_nvfp4 and self.support_nvfp4_output and enable_attn_nvfp4_output and not self.attn_output_gate:
             out_scale_sf = self.o_proj.input_scale
 
         kv_scales_sf = None
@@ -751,7 +740,7 @@ class MLA(nn.Module):
             self.register_to_config = True
 
         # only support one kind of sparse attention, dsa now.
-        if config.sparse_attention_config is not None:
+        if config is not None and config.sparse_attention_config is not None:
             self.is_dsa = True
         else:
             self.is_dsa = False
@@ -963,8 +952,6 @@ class MLA(nn.Module):
         if not config.skip_create_weights_in_init:
             self.create_weights()
 
-        self.indexer = self.mqa.indexer if self.is_dsa else None
-
     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__
@@ -1198,7 +1185,7 @@ class MLA(nn.Module):
                 assert position_ids is not None
                 k_pe_gen = self.apply_rope(q_gen, k_pe_gen, position_ids)
 
-            self.forward_absorption(
+            self.forward_absorption_generation(
                 q_gen,
                 compressed_kv_gen,
                 k_pe_gen,
@@ -1234,9 +1221,21 @@ class MLA(nn.Module):
         if position_ids is not None:
             position_ids = position_ids[..., :num_tokens]
 
-        q, compressed_kv, k_pe = self.kv_a_proj_with_mqa(hidden_states).split(
-            [self.q_lora_rank, self.kv_lora_rank, self.qk_rope_head_dim], -1)
+        if self.fuse_a_indexer_k_weight:
+            q, compressed_kv, k_pe, indexer_k, indexer_weights = self.kv_a_proj_with_mqa(
+                hidden_states).split([
+                    self.q_lora_rank, self.kv_lora_rank, self.qk_rope_head_dim,
+                    self.indexer.head_dim, self.indexer.n_heads
+                ], -1)
+        else:
+            q, compressed_kv, k_pe = self.kv_a_proj_with_mqa(
+                hidden_states).split([
+                    self.q_lora_rank, self.kv_lora_rank, self.qk_rope_head_dim
+                ], -1)
+            indexer_k = None
+            indexer_weights = None
 
+        # TODO: possibly overlap/fuse q_a_rmsnorm + kv_a_rmsnorm + indexer.k_layernorm?
         q, compressed_kv = maybe_execute_in_parallel(
             lambda: self.q_a_layernorm(q),
             lambda: self.kv_a_layernorm(compressed_kv),
@@ -1245,12 +1244,18 @@ class MLA(nn.Module):
             self.aux_stream,
         )
         qr = q
-        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,
+        latent_cache = torch.concat([compressed_kv, k_pe], dim=-1)
+
+        # TODO: fuse wq_b + (indexer) wlq here
+        q = self.q_b_proj(q)
+        # Indexer
+        topk_indices = self.indexer(
+            qr,
+            hidden_states,
+            attn_metadata,
+            position_ids,
+            indexer_k=indexer_k,  # indexer K proj
+            indexer_weights=indexer_weights,  # indexer weights proj
         )
 
         assert q.shape[
@@ -1258,10 +1263,6 @@ class MLA(nn.Module):
 
         assert output is not None, "output must be provided"
 
-        # Indexer
-        topk_indices = self.indexer(qr, hidden_states, attn_metadata,
-                                    position_ids)
-
         if num_contexts > 0:
             q_ctx = q[:num_ctx_tokens, ...]
             compressed_kv_ctx = compressed_kv[:num_ctx_tokens, ...]
@@ -1355,14 +1356,13 @@ class MLA(nn.Module):
         topk_indices: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
         if get_sm_version() >= 100:
-            return self.forward_absorption(q,
-                                           compressed_kv,
-                                           k_pe,
-                                           attn_metadata,
-                                           output,
-                                           latent_cache=latent_cache,
-                                           topk_indices=topk_indices,
-                                           is_generation=False)
+            return self.forward_absorption_context(q,
+                                                   compressed_kv,
+                                                   k_pe,
+                                                   attn_metadata,
+                                                   output,
+                                                   latent_cache=latent_cache,
+                                                   topk_indices=topk_indices)
         else:
             return self.forward_sparse_mla_kvcache_bf16(q,
                                                         latent_cache,
@@ -1382,13 +1382,13 @@ class MLA(nn.Module):
         topk_indices: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
         if get_sm_version() >= 100:
-            return self.forward_absorption(q,
-                                           compressed_kv,
-                                           k_pe,
-                                           attn_metadata,
-                                           output,
-                                           latent_cache=latent_cache,
-                                           topk_indices=topk_indices)
+            return self.forward_absorption_generation(q,
+                                                      compressed_kv,
+                                                      k_pe,
+                                                      attn_metadata,
+                                                      output,
+                                                      latent_cache=latent_cache,
+                                                      topk_indices=topk_indices)
         else:
             return self.forward_sparse_mla_kvcache_bf16(q,
                                                         latent_cache,
@@ -1653,7 +1653,172 @@ class MLA(nn.Module):
                                             position_ids, attn_metadata, output,
                                             latent_cache)
 
-    def forward_absorption(
+    def forward_absorption_generation(
+        self,
+        q: torch.Tensor,
+        compressed_kv: torch.Tensor,
+        k_pe: torch.Tensor,
+        attn_metadata: AttentionMetadata,
+        output: torch.Tensor,
+        position_ids: Optional[torch.Tensor] = None,
+        latent_cache: Optional[torch.Tensor] = None,
+        topk_indices: Optional[torch.Tensor] = None,
+    ) -> torch.Tensor:
+        num_tokens = q.shape[0]
+        q_nope, q_pe = q.view([-1, self.num_heads_tp, 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
+        num_seqs = attn_metadata.kv_lens_cuda_runtime.size(0)
+
+        cu_q_seqlens = torch.empty(num_seqs + 1,
+                                   dtype=torch.int32,
+                                   device=q.device)
+        cu_kv_seqlens = torch.empty(num_seqs + 1,
+                                    dtype=torch.int32,
+                                    device=q.device)
+        fmha_scheduler_counter = torch.empty(1,
+                                             dtype=torch.uint32,
+                                             device=q.device)
+        has_fp8_kv_cache = self.mqa.has_fp8_kv_cache if hasattr(
+            self.mqa, 'has_fp8_kv_cache') else False
+
+        mla_bmm1_scale = None
+        mla_bmm2_scale = None
+        quant_q_buffer = None
+        if has_fp8_kv_cache:
+            mla_bmm1_scale = torch.empty(2,
+                                         dtype=torch.float32,
+                                         device=q.device)
+            mla_bmm2_scale = torch.empty(1,
+                                         dtype=torch.float32,
+                                         device=q.device)
+            quant_q_buffer = torch.empty(
+                num_tokens,
+                self.num_heads_tp, (self.kv_lora_rank + self.qk_rope_head_dim),
+                dtype=torch.uint8,
+                device=q.device)
+
+        fused_q = torch.empty(
+            [
+                num_tokens, self.num_heads_tp,
+                (self.kv_lora_rank + self.qk_rope_head_dim)
+            ],
+            dtype=q.dtype,
+            device=q.device,
+        )
+
+        rope_stream = self.aux_stream if not has_fp8_kv_cache else None
+        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
+            maybe_execute_in_parallel(
+                lambda: torch.ops.trtllm.bmm_out(
+                    q_nope_t, self.k_b_proj_trans.transpose(1, 2), q_nope_out),
+                lambda: self.mqa.mla_rope_generation(
+                    fused_q, q_pe, latent_cache, attn_metadata, cu_q_seqlens,
+                    cu_kv_seqlens, fmha_scheduler_counter, mla_bmm1_scale,
+                    mla_bmm2_scale, quant_q_buffer),
+                self.ln_events[0],
+                self.ln_events[1],
+                rope_stream,
+            )
+
+        elif self.k_b_proj_trans.dtype == torch.float8_e4m3fn:
+            # [num_heads, num_tokens, self.kv_lora_rank]
+            q_nope_out = fused_q[..., :self.kv_lora_rank].transpose(0, 1)
+
+            maybe_execute_in_parallel(
+                lambda: fp8_block_scaling_bmm_out(
+                    q_nope,
+                    self.k_b_proj_trans,
+                    self.k_b_proj_trans_scale,
+                    q_nope_out,
+                    self.k_b_proj_trans_dequant,
+                ),
+                lambda: self.mqa.mla_rope_generation(
+                    fused_q, q_pe, latent_cache, attn_metadata, cu_q_seqlens,
+                    cu_kv_seqlens, fmha_scheduler_counter, mla_bmm1_scale,
+                    mla_bmm2_scale, quant_q_buffer),
+                self.ln_events[0],
+                self.ln_events[1],
+                rope_stream,
+            )
+        else:
+            raise NotImplementedError(
+                f"Missing bmm impl for dtype: {self.k_b_proj_trans.dtype}.")
+
+        fused_q = fused_q.view([
+            num_tokens,
+            self.num_heads_tp * (self.kv_lora_rank + self.qk_rope_head_dim)
+        ])
+
+        # Use generation_only for generation phase and context_only for context phase in DSA attention
+        attention_input_type = AttentionInputType.generation_only
+
+        attn_out_latent = self._attn_forward_gen(
+            self.mqa,
+            fused_q,
+            None,
+            None,
+            position_ids,
+            attn_metadata,
+            attention_input_type=attention_input_type,
+            out_scale=self.out_scale,
+            latent_cache=latent_cache,  # kvcache and k_pe
+            q_pe=q_pe,  # used by `invokeMLARopeGeneration`
+            topk_indices=topk_indices,  # used by DSA attention
+            is_generation=True,  # used by DSA attention
+            cu_q_seqlens=cu_q_seqlens,  # used by `mlaGeneration`
+            cu_kv_seqlens=cu_kv_seqlens,  # used by `mlaGeneration`
+            fmha_scheduler_counter=
+            fmha_scheduler_counter,  # used by `mlaGeneration`
+            mla_bmm1_scale=mla_bmm1_scale,  # used by `mlaGeneration`
+            mla_bmm2_scale=mla_bmm2_scale,  # used by `mlaGeneration`
+            quant_q_buffer=quant_q_buffer,  # used by `mlaGeneration`
+        )
+        fused_q = None
+
+        # note: if we do not have CP, then num_heads_tp_cp == num_heads_tp
+        assert (attn_out_latent.shape[0] == q.shape[0]
+                and attn_out_latent.shape[1]
+                == self.num_heads_tp_cp * self.kv_lora_rank)
+
+        # [seq, num_heads, kv_lora_rank]
+        attn_out_latent = attn_out_latent.view(
+            [-1, self.num_heads_tp_cp, self.kv_lora_rank])
+
+        attn_output = output.view(
+            [num_tokens, self.num_heads_tp_cp, self.v_head_dim])
+
+        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:
+            fp8_block_scaling_bmm_out(
+                attn_out_latent,
+                self.v_b_proj,
+                self.v_b_proj_scale,
+                attn_output.transpose(0, 1),
+                self.v_b_proj_dequant,
+            )
+        else:
+            raise NotImplementedError(
+                f"Missing bmm impl for dtype: {self.v_b_proj.dtype}.")
+
+        return output
+
+    def forward_absorption_context(
         self,
         q: torch.Tensor,
         compressed_kv: torch.Tensor,
@@ -1663,7 +1828,6 @@ class MLA(nn.Module):
         position_ids: Optional[torch.Tensor] = None,
         latent_cache: Optional[torch.Tensor] = None,
         topk_indices: Optional[torch.Tensor] = None,
-        is_generation: bool = True,
     ) -> torch.Tensor:
         num_tokens = q.shape[0]
         q_nope, q_pe = q.view([-1, self.num_heads_tp, self.qk_head_dim]).split(
@@ -1715,7 +1879,7 @@ class MLA(nn.Module):
         ])
 
         # Use generation_only for generation phase and context_only for context phase in DSA attention
-        attention_input_type = AttentionInputType.generation_only if is_generation else AttentionInputType.context_only
+        attention_input_type = AttentionInputType.context_only
         attn_out_latent = self._attn_forward_gen(
             self.mqa,
             fused_q,
@@ -1728,7 +1892,7 @@ class MLA(nn.Module):
             latent_cache=latent_cache,  # kvcache and k_pe
             q_pe=q_pe,  # used by `invokeMLARopeGeneration`
             topk_indices=topk_indices,  # used by DSA attention
-            is_generation=is_generation,  # used by DSA attention
+            is_generation=False,  # used by DSA attention
         )
         fused_q = None
 
@@ -1794,10 +1958,10 @@ class MLA(nn.Module):
                 q, latent_cache, attn_metadata, is_generation=is_generation)
 
         num_tokens = q.shape[0]
-        q_nope, q_rope = q.view(-1, self.num_heads, self.qk_head_dim).split(
+        q_nope, q_rope = q.view(-1, self.num_heads_tp, self.qk_head_dim).split(
             [self.qk_nope_head_dim, self.qk_rope_head_dim], dim=-1)
         q_nope_out = torch.empty(
-            [num_tokens, self.num_heads, (self.kv_lora_rank)],
+            [num_tokens, self.num_heads_tp, (self.kv_lora_rank)],
             dtype=q.dtype,
             device=q.device,
         )
@@ -1836,23 +2000,23 @@ class MLA(nn.Module):
         # FlashMLA sparse kernel (bf16) requires num_heads=128 on sm100 or multiple of 64 on sm90
         if sm_version >= 100:
             padding = 128
-            assert self.num_heads <= padding, (
+            assert self.num_heads_tp <= padding, (
                 f"SM100 FlashMLA sparse kernel requires exactly {padding} heads, "
-                f"got {self.num_heads}. Padding from values > {padding} is not supported."
+                f"got {self.num_heads_tp}. Padding from values > {padding} is not supported."
             )
         else:  # SM90
-            padding = ((self.num_heads + 63) // 64) * 64  # multiple of 64
+            padding = ((self.num_heads_tp + 63) // 64) * 64  # multiple of 64
 
-        if self.num_heads != padding:
+        if self.num_heads_tp != padding:
             logger.warning_once(
-                f"Padding num_heads from {self.num_heads} to {padding} "
+                f"Padding num_heads from {self.num_heads_tp} to {padding} "
                 f"due to FlashMLA sparse attention kernel requirement",
                 key="sparse_mla_padding_warning")
 
             # Create padded tensor with zeros for extra heads
             q_padded = q_concat.new_empty(
                 (num_tokens, padding, q_concat.shape[2]))
-            q_padded[:, :self.num_heads, :] = q_concat
+            q_padded[:, :self.num_heads_tp, :] = q_concat
             q_concat = q_padded
 
         # Convert indices and return all-layer KV pool
@@ -1874,17 +2038,17 @@ class MLA(nn.Module):
                 "flash_mla_sparse_fwd not available. Please ensure FlashMLA module is built."
             )
 
-        # [seq, num_heads, kv_lora_rank]
-        attn_out_latent = attn_out_latent[:, :self.
-                                          num_heads, :]  # account for padding
+        # [seq, num_heads, kv_lora_rank], account for padding
+        attn_out_latent = attn_out_latent[:, :self.num_heads_tp, :]
         # TODO: seems we need .contiguous() here when padding enabled before pass to bmm?
         attn_out_latent = attn_out_latent.view(
-            [-1, self.num_heads, self.kv_lora_rank])
+            [-1, self.num_heads_tp, self.kv_lora_rank])
 
         assert (attn_out_latent.shape[0] == q.shape[0]
-                and attn_out_latent.shape[1] == self.num_heads)
+                and attn_out_latent.shape[1] == self.num_heads_tp)
 
-        attn_output = output.view([num_tokens, self.num_heads, self.v_head_dim])
+        attn_output = output.view(
+            [num_tokens, self.num_heads_tp, self.v_head_dim])
 
         if self.v_b_proj.dtype == torch.bfloat16:
             # [num_heads, seq, kv_lora_rank] x [num_heads, kv_lora_rank, v_head_dim]

latest/_downloads/c68095123d889975e6e5e839a4241d22/model_engine.py

@@ -54,13 +54,12 @@ from ..speculative.utils import SpecDecodingTensor
 from ..utils import (get_model_extra_attrs,
                      set_per_request_piecewise_cuda_graph_flag,
                      set_torch_compiling, with_model_extra_attrs)
-from .config import _construct_checkpoint_loader
 from .config_utils import is_mla
-from .cuda_graph_runner import CUDAGraphRunner
+from .cuda_graph_runner import CUDAGraphRunner, CUDAGraphRunnerConfig
 from .guided_decoder import CapturableGuidedDecoder
 from .layerwise_nvtx_marker import LayerwiseNvtxMarker
-from .llm_request import get_draft_token_length
-from .model_loader import ModelLoader
+from .llm_request import LlmRequest, get_draft_token_length
+from .model_loader import ModelLoader, _construct_checkpoint_loader
 from .resource_manager import (BaseResourceManager, KVCacheManager,
                                ResourceManager, ResourceManagerType)
 from .sampler import SampleStateTensors
@@ -74,14 +73,13 @@ class ModelEngine(ABC):
         raise NotImplementedError
 
     @abstractmethod
-    def forward(
-        self,
-        scheduled_requests: ScheduledRequests,
-        resource_manager: ResourceManager,
-        new_tensors_device: Optional[SampleStateTensors],
-        gather_context_logits: bool = False,
-        cache_indirection_buffer: Optional[torch.Tensor] = None,
-    ):
+    def forward(self,
+                scheduled_requests: ScheduledRequests,
+                resource_manager: ResourceManager,
+                new_tensors_device: Optional[SampleStateTensors],
+                gather_context_logits: bool = False,
+                cache_indirection_buffer: Optional[torch.Tensor] = None,
+                num_accepted_tokens_device: Optional[torch.Tensor] = None):
         raise NotImplementedError
 
     def warmup(self, resource_manager: ResourceManager) -> None:
@@ -366,14 +364,60 @@ class PyTorchModelEngine(ModelEngine):
         if self.use_mrope:
             self.mrope_position_ids_cuda = torch.empty(
                 (3, 1, self.max_num_tokens), dtype=torch.int, device='cuda')
-        self.iter_counter = 0
+
+        # Pre-allocated buffers for draft model to avoid implicit synchronization
+        # These are used to build index tensors without creating tensors from Python lists
+        if is_draft_model:
+            # Buffers for context and first_draft input_ids updates
+            self.draft_ctx_token_indices_cuda = torch.empty((self.batch_size, ),
+                                                            dtype=torch.long,
+                                                            device='cuda')
+            self.draft_ctx_seq_slots_cuda = torch.empty((self.batch_size, ),
+                                                        dtype=torch.long,
+                                                        device='cuda')
+            # Buffers for first_draft requests (max_draft_len+1 tokens per request)
+            max_first_draft_tokens = self.batch_size * (
+                self.original_max_draft_len +
+                1) if spec_config else self.batch_size
+            self.draft_first_draft_indices_cuda = torch.empty(
+                (max_first_draft_tokens, ), dtype=torch.long, device='cuda')
+            self.draft_first_draft_seq_slots_cuda = torch.empty(
+                (max_first_draft_tokens, ), dtype=torch.long, device='cuda')
+            # Buffers for seq_slots and request indices
+            self.draft_seq_slots_buffer_cuda = torch.empty((self.batch_size, ),
+                                                           dtype=torch.int,
+                                                           device='cuda')
+            self.draft_request_indices_buffer_cuda = torch.empty(
+                (self.batch_size, ), dtype=torch.int, device='cuda')
 
         # We look up this key in resource_manager during forward to find the
         # kv cache manager. Can be changed to support multiple model engines
         # with different KV cache managers.
-        self.kv_cache_manager_key = ResourceManagerType.KV_CACHE_MANAGER
+        self.kv_cache_manager_key = ResourceManagerType.DRAFT_KV_CACHE_MANAGER if is_draft_model else ResourceManagerType.KV_CACHE_MANAGER
         self.lora_model_config: Optional[LoraModelConfig] = None
-        self.cuda_graph_runner = CUDAGraphRunner(self)
+
+        # Create config and runner
+        cuda_graph_runner_config = CUDAGraphRunnerConfig(
+            use_cuda_graph=self.cuda_graph_config is not None,
+            cuda_graph_padding_enabled=self._cuda_graph_padding_enabled,
+            cuda_graph_batch_sizes=self._cuda_graph_batch_sizes,
+            max_cuda_graph_batch_size=self._max_cuda_graph_batch_size,
+            max_beam_width=self.max_beam_width,
+            spec_config=self.spec_config,
+            cuda_graph_mem_pool=self._cuda_graph_mem_pool,
+            max_num_tokens=self.max_num_tokens,
+            use_mrope=self.use_mrope,
+            original_max_draft_len=self.original_max_draft_len,
+            original_max_total_draft_tokens=self.
+            original_max_total_draft_tokens,
+            is_draft_model=self.is_draft_model,
+            enable_attention_dp=self.enable_attention_dp,
+            batch_size=self.batch_size,
+            mapping=self.mapping,
+            dist=self.dist,
+            kv_cache_manager_key=self.kv_cache_manager_key,
+        )
+        self.cuda_graph_runner = CUDAGraphRunner(cuda_graph_runner_config)
 
         # Setup the local cache indirection buffer only once and reuse it.
         # This way it can also be used for CUDA graphs.
@@ -587,6 +631,13 @@ class PyTorchModelEngine(ModelEngine):
             with self._release_batch_context(warmup_request,
                                              resource_manager) as batch:
                 if batch is not None:
+                    # Reset the flag is_first_draft for the draft model.
+                    # This is necessary for overlap scheduler.
+                    spec_resource_manager = resource_manager.get_resource_manager(
+                        ResourceManagerType.SPEC_RESOURCE_MANAGER)
+                    if self.is_draft_model and isinstance(
+                            spec_resource_manager, Eagle3ResourceManager):
+                        spec_resource_manager.is_first_draft = True
                     self.forward(batch,
                                  new_tensors_device=None,
                                  resource_manager=resource_manager)
@@ -909,6 +960,18 @@ class PyTorchModelEngine(ModelEngine):
                 self.attn_runtime_features.cache_reuse
                 or self.attn_runtime_features.chunked_prefill)
         cache_indirection = self.cache_indirection_attention if self.attn_backend.Metadata is TrtllmAttentionMetadata else None
+        num_attention_heads = getattr(self.model.model_config.pretrained_config,
+                                      'num_attention_heads', None)
+        if num_attention_heads is not None:
+            num_key_value_heads = getattr(
+                self.model.model_config.pretrained_config,
+                'num_key_value_heads', None)
+            if num_key_value_heads is not None:
+                num_heads_per_kv = num_attention_heads // num_key_value_heads
+            else:
+                num_heads_per_kv = 1
+        else:
+            num_heads_per_kv = 1
         if kv_cache_manager is None:
             return self.attn_backend.Metadata(
                 max_num_requests=self.batch_size,
@@ -921,7 +984,8 @@ class PyTorchModelEngine(ModelEngine):
                 enable_context_mla_with_cached_kv=
                 enable_context_mla_with_cached_kv,
                 cache_indirection=cache_indirection,
-                sparse_attention_config=self.sparse_attention_config)
+                sparse_attention_config=self.sparse_attention_config,
+                num_heads_per_kv=num_heads_per_kv)
 
         if self.attn_metadata is not None:
             # This assertion can be relaxed if needed: just create a new metadata
@@ -939,7 +1003,9 @@ class PyTorchModelEngine(ModelEngine):
             enable_flash_mla=self.model.model_config.enable_flash_mla,
             enable_context_mla_with_cached_kv=enable_context_mla_with_cached_kv,
             cache_indirection=cache_indirection,
-            sparse_attention_config=self.sparse_attention_config)
+            sparse_attention_config=self.sparse_attention_config,
+            num_heads_per_kv=num_heads_per_kv,
+        )
 
         return self.attn_metadata
 
@@ -1224,10 +1290,13 @@ class PyTorchModelEngine(ModelEngine):
             attn_metadata: AttentionMetadata,
             spec_metadata: Optional[SpecMetadata] = None,
             new_tensors_device: Optional[SampleStateTensors] = None,
-            cache_indirection_buffer: Optional[torch.Tensor] = None):
+            cache_indirection_buffer: Optional[torch.Tensor] = None,
+            num_accepted_tokens_device: Optional[torch.Tensor] = None,
+            req_id_to_old_request: Optional[Dict[int, LlmRequest]] = None):
         """
         Prepare inputs for Pytorch Model.
         """
+
         new_tokens_device, new_tokens_lens_device, next_draft_tokens_device = None, None, None
         if new_tensors_device is not None:
             # speculative decoding cases: [batch, 1 + draft_len], others: [batch]
@@ -1258,6 +1327,19 @@ class PyTorchModelEngine(ModelEngine):
         mrope_position_ids = []
         num_accepted_draft_tokens = []  # per request
 
+        # Variables for updating the inputs of draft model
+        # Base values for gather_ids computation
+        first_draft_base_gather_ids = []
+        # seq_slots to index into num_accepted_tokens_device
+        first_draft_seq_slots = []
+        # Indices in the num_accepted_draft_tokens list
+        first_draft_request_indices = []
+
+        # (start_idx, end_idx, seq_slot) for context requests
+        context_input_ids_positions = []
+        # (start_idx, end_idx, seq_slot) for first_draft requests
+        first_draft_input_ids_positions = []
+
         for request in scheduled_requests.context_requests:
             request_ids.append(request.py_request_id)
             all_prompt_tokens = request.get_tokens(0)
@@ -1267,7 +1349,20 @@ class PyTorchModelEngine(ModelEngine):
             prompt_tokens = all_prompt_tokens[begin_compute:end_compute]
             position_ids.extend(
                 range(begin_compute, begin_compute + len(prompt_tokens)))
-            input_ids.extend(prompt_tokens)
+
+            # Track position for updating the inputs of draft model
+            if self.is_draft_model and num_accepted_tokens_device is not None:
+                start_idx = len(input_ids)
+                input_ids.extend(prompt_tokens)
+                end_idx = len(input_ids)
+                slot_idx = req_id_to_old_request[
+                    request.py_request_id].py_seq_slot
+                context_input_ids_positions.append(
+                    (start_idx, end_idx - 1,
+                     slot_idx))  # end_idx-1 is the last token position
+            else:
+                input_ids.extend(prompt_tokens)
+
             gather_ids.append(len(input_ids) - 1)
             sequence_lengths.append(len(prompt_tokens))
             num_accepted_draft_tokens.append(len(prompt_tokens) - 1)
@@ -1411,14 +1506,9 @@ class PyTorchModelEngine(ModelEngine):
                 previous_batch_indices.append(previous_batch_idx)
                 previous_pos_indices.extend([previous_batch_idx] *
                                             (1 + self.runtime_draft_len))
-                if self.spec_config.spec_dec_mode.has_draft_model():
-                    # In the overlap scheduler workflow, if having draft model, we already updated the previous batch before launching the target model,
-                    # so we only need to add the runtime_draft_len to the past_seen_token_num.
-                    num_cached_tokens_per_seq.append(past_seen_token_num +
-                                                     self.runtime_draft_len)
-                else:
-                    num_cached_tokens_per_seq.append(past_seen_token_num +
-                                                     self.runtime_draft_len + 1)
+
+                num_cached_tokens_per_seq.append(past_seen_token_num +
+                                                 self.runtime_draft_len + 1)
                 request.cached_tokens = num_cached_tokens_per_seq[-1]
                 if self.enable_spec_decode and spec_config.spec_dec_mode.extend_ctx(
                         self.attn_backend):
@@ -1436,13 +1526,39 @@ class PyTorchModelEngine(ModelEngine):
             prompt_tokens = all_prompt_tokens[begin_compute:end_compute]
             position_ids.extend(
                 range(begin_compute, begin_compute + len(prompt_tokens)))
-            input_ids.extend(prompt_tokens)
-            gather_ids.append(
-                len(input_ids) - 1 - (self.original_max_draft_len -
-                                      request.py_num_accepted_draft_tokens))
+
+            # Track position for updating the inputs of draft model
+            if self.is_draft_model and num_accepted_tokens_device is not None:
+                start_idx = len(input_ids)
+                input_ids.extend(prompt_tokens)
+                end_idx = len(input_ids)
+                # For first_draft, we need to replace the last original_max_draft_len+1 tokens
+                slot_idx = req_id_to_old_request[
+                    request.py_request_id].py_seq_slot
+                first_draft_input_ids_positions.append(
+                    (start_idx, end_idx, slot_idx))
+
+                # Store info for GPU computation of gather_ids and num_accepted_draft_tokens
+                base_gather_id = len(
+                    input_ids) - 1 - self.original_max_draft_len
+                # Placeholder, will be corrected on GPU
+                gather_ids.append(base_gather_id)
+                first_draft_base_gather_ids.append(base_gather_id)
+                first_draft_seq_slots.append(slot_idx)
+                first_draft_request_indices.append(
+                    len(num_accepted_draft_tokens))
+
+                # Placeholder, will be corrected on GPU
+                num_accepted_draft_tokens.append(0)
+            else:
+                input_ids.extend(prompt_tokens)
+                gather_ids.append(
+                    len(input_ids) - 1 - (self.original_max_draft_len -
+                                          request.py_num_accepted_draft_tokens))
+                num_accepted_draft_tokens.append(
+                    request.py_num_accepted_draft_tokens)
+
             sequence_lengths.append(1 + self.original_max_draft_len)
-            num_accepted_draft_tokens.append(
-                request.py_num_accepted_draft_tokens)
             prompt_lengths.append(request.py_prompt_len)
             past_seen_token_num = begin_compute
             num_cached_tokens_per_seq.append(past_seen_token_num)
@@ -1462,7 +1578,17 @@ class PyTorchModelEngine(ModelEngine):
                     # skip adding input_ids of CUDA graph dummy requests so that new_tokens_device
                     # can be aligned to the correct positions.
                     if not request.is_cuda_graph_dummy:
-                        input_ids.append(request.get_last_tokens(beam))
+                        # Track position for GPU update (draft model only)
+                        if self.is_draft_model and num_accepted_tokens_device is not None:
+                            start_idx = len(input_ids)
+                            input_ids.append(request.get_last_tokens(beam))
+                            end_idx = len(input_ids)
+                            slot_idx = req_id_to_old_request[
+                                request.py_request_id].py_seq_slot
+                            first_draft_input_ids_positions.append(
+                                (start_idx, end_idx, slot_idx))
+                        else:
+                            input_ids.append(request.get_last_tokens(beam))
                     past_seen_token_num = request.max_beam_num_tokens - 1
                 else:
                     # the request has previous tensor
@@ -1537,6 +1663,79 @@ class PyTorchModelEngine(ModelEngine):
                                      dtype=torch.int,
                                      pin_memory=True)
             self.input_ids_cuda[:num_tokens].copy_(input_ids, non_blocking=True)
+
+            # Update input_ids_cuda with new tokens from new_tensors_device (draft model only)
+            if self.is_draft_model and num_accepted_tokens_device is not None:
+                # For context requests: replace the last token with new_tensors_device[0, seq_slot, 0]
+                if len(context_input_ids_positions) > 0:
+                    # Build tensors on CPU first, then copy to GPU to avoid implicit sync
+                    num_ctx_positions = len(context_input_ids_positions)
+                    ctx_token_indices_cpu = torch.tensor([
+                        last_token_idx
+                        for _, last_token_idx, _ in context_input_ids_positions
+                    ],
+                                                         dtype=torch.long,
+                                                         pin_memory=True)
+                    ctx_seq_slots_cpu = torch.tensor([
+                        seq_slot
+                        for _, _, seq_slot in context_input_ids_positions
+                    ],
+                                                     dtype=torch.long,
+                                                     pin_memory=True)
+                    # Copy to pre-allocated GPU buffers
+                    self.draft_ctx_token_indices_cuda[:num_ctx_positions].copy_(
+                        ctx_token_indices_cpu, non_blocking=True)
+                    self.draft_ctx_seq_slots_cuda[:num_ctx_positions].copy_(
+                        ctx_seq_slots_cpu, non_blocking=True)
+                    self.input_ids_cuda[
+                        self.
+                        draft_ctx_token_indices_cuda[:num_ctx_positions]] = new_tensors_device.new_tokens[
+                            0,
+                            self.draft_ctx_seq_slots_cuda[:num_ctx_positions],
+                            0]
+
+                # For first_draft requests: replace the last (original_max_draft_len+1) tokens
+                # with new_tensors_device[:, seq_slot, 0]
+                if len(first_draft_input_ids_positions) > 0:
+                    # All first_draft requests have same token length (original_max_draft_len + 1)
+                    # Build index tensors on CPU first, then copy to GPU to avoid implicit sync
+                    num_requests = len(first_draft_input_ids_positions)
+                    tokens_per_request = first_draft_input_ids_positions[0][
+                        1] - first_draft_input_ids_positions[0][0]
+
+                    # Create flat index array for all tokens to update on CPU
+                    all_indices = []
+                    all_seq_slots = []
+                    for start_idx, end_idx, seq_slot in first_draft_input_ids_positions:
+                        all_indices.extend(range(start_idx, end_idx))
+                        all_seq_slots.extend([seq_slot] * (end_idx - start_idx))
+
+                    # Create CPU tensors with pinned memory
+                    total_tokens = len(all_indices)
+                    idx_tensor_cpu = torch.tensor(all_indices,
+                                                  dtype=torch.long,
+                                                  pin_memory=True)
+                    seq_slots_tensor_cpu = torch.tensor(all_seq_slots,
+                                                        dtype=torch.long,
+                                                        pin_memory=True)
+
+                    # Copy to pre-allocated GPU buffers
+                    self.draft_first_draft_indices_cuda[:total_tokens].copy_(
+                        idx_tensor_cpu, non_blocking=True)
+                    self.draft_first_draft_seq_slots_cuda[:total_tokens].copy_(
+                        seq_slots_tensor_cpu, non_blocking=True)
+
+                    # Create token position indices (repeating 0..tokens_per_request for each request)
+                    token_positions = torch.arange(
+                        tokens_per_request, dtype=torch.long,
+                        device='cuda').repeat(num_requests)
+
+                    self.input_ids_cuda[
+                        self.
+                        draft_first_draft_indices_cuda[:total_tokens]] = new_tensors_device.new_tokens[
+                            token_positions, self.
+                            draft_first_draft_seq_slots_cuda[:total_tokens], 0]
+
         if num_draft_tokens > 0:
             draft_tokens = torch.tensor(draft_tokens,
                                         dtype=torch.int,
@@ -1550,6 +1749,33 @@ class PyTorchModelEngine(ModelEngine):
             self.num_accepted_draft_tokens_cuda[:len(
                 num_accepted_draft_tokens)].copy_(num_accepted_draft_tokens,
                                                   non_blocking=True)
+
+            # Update num_accepted_draft_tokens_cuda for first_draft_requests directly from num_accepted_tokens_device (draft model only)
+            if self.is_draft_model and len(first_draft_seq_slots) > 0:
+                # Build tensors on CPU first, then copy to GPU to avoid implicit sync
+                num_first_draft = len(first_draft_seq_slots)
+                first_draft_seq_slots_cpu = torch.tensor(first_draft_seq_slots,
+                                                         dtype=torch.int,
+                                                         pin_memory=True)
+                first_draft_indices_cpu = torch.tensor(
+                    first_draft_request_indices,
+                    dtype=torch.int,
+                    pin_memory=True)
+
+                # Copy to pre-allocated GPU buffers
+                self.draft_seq_slots_buffer_cuda[:num_first_draft].copy_(
+                    first_draft_seq_slots_cpu, non_blocking=True)
+                self.draft_request_indices_buffer_cuda[:num_first_draft].copy_(
+                    first_draft_indices_cpu, non_blocking=True)
+
+                # Extract accepted tokens for first_draft requests from device tensor
+                accepted_tokens = num_accepted_tokens_device[
+                    self.draft_seq_slots_buffer_cuda[:num_first_draft]]
+                # Update the correct positions in num_accepted_draft_tokens_cuda
+                self.num_accepted_draft_tokens_cuda[
+                    self.
+                    draft_request_indices_buffer_cuda[:
+                                                      num_first_draft]] = accepted_tokens
         if next_draft_tokens_device is not None:
             # Initialize these two values to zeros
             self.previous_pos_id_offsets_cuda *= 0
@@ -1655,6 +1881,34 @@ class PyTorchModelEngine(ModelEngine):
                 gather_ids, dtype=torch.int, pin_memory=True),
                                                          non_blocking=True)
 
+            # Update gather_ids for first_draft_requests on GPU (draft model only)
+            if self.is_draft_model and len(first_draft_seq_slots) > 0:
+                # Build tensors on CPU first, then copy to GPU to avoid implicit sync
+                num_first_draft = len(first_draft_seq_slots)
+                first_draft_seq_slots_cpu = torch.tensor(first_draft_seq_slots,
+                                                         dtype=torch.int,
+                                                         pin_memory=True)
+                first_draft_indices_cpu = torch.tensor(
+                    first_draft_request_indices,
+                    dtype=torch.int,
+                    pin_memory=True)
+
+                # Copy to pre-allocated GPU buffers
+                self.draft_seq_slots_buffer_cuda[:num_first_draft].copy_(
+                    first_draft_seq_slots_cpu, non_blocking=True)
+                self.draft_request_indices_buffer_cuda[:num_first_draft].copy_(
+                    first_draft_indices_cpu, non_blocking=True)
+
+                # Extract accepted tokens for first_draft requests from device tensor
+                accepted_tokens = num_accepted_tokens_device[
+                    self.draft_seq_slots_buffer_cuda[:num_first_draft]]
+                # Update gather_ids: gather_id = base_gather_id + num_accepted_tokens
+                # (since gather_id = len(input_ids) - 1 - (max_draft_len - num_accepted))
+                self.gather_ids_cuda[
+                    self.
+                    draft_request_indices_buffer_cuda[:
+                                                      num_first_draft]] += accepted_tokens
+
         if not attn_metadata.is_cuda_graph:
             # Assumes seq lens do not change between CUDA graph invocations. This applies
             # to draft sequences too. This means that all draft sequences must be padded.
@@ -1671,8 +1925,11 @@ class PyTorchModelEngine(ModelEngine):
             is_cuda_graph_during_warmup = self.is_warmup and attn_metadata.is_cuda_graph
             if cache_indirection_buffer is not None:
                 #Copy cache indirection to local buffer with offsets changing:  seq_slots[i] -> i
+                # Convert to GPU tensor to avoid implicit sync
+                gen_request_seq_slots_tensor = torch.tensor(
+                    gen_request_seq_slots, dtype=torch.long, device='cuda')
                 self.cache_indirection_attention[:num_generation_requests].copy_(
-                    cache_indirection_buffer[gen_request_seq_slots])
+                    cache_indirection_buffer[gen_request_seq_slots_tensor])
             if cache_indirection_buffer is not None or is_cuda_graph_during_warmup:
                 attn_metadata.beam_width = self.max_beam_width
         else:
@@ -2253,7 +2510,9 @@ class PyTorchModelEngine(ModelEngine):
             attn_metadata: AttentionMetadata,
             spec_metadata: Optional[SpecMetadata] = None,
             new_tensors_device: Optional[SampleStateTensors] = None,
-            cache_indirection_buffer: Optional[torch.Tensor] = None):
+            cache_indirection_buffer: Optional[torch.Tensor] = None,
+            num_accepted_tokens_device: Optional[torch.Tensor] = None,
+            req_id_to_old_request: Optional[Dict[int, LlmRequest]] = None):
         if self.mapping is not None and 'cp_type' in self.mapping.cp_config:
             cp_type = self.mapping.cp_config['cp_type']
             if CpType.STAR == cp_type:
@@ -2269,19 +2528,21 @@ class PyTorchModelEngine(ModelEngine):
         return self._prepare_tp_inputs(scheduled_requests, kv_cache_manager,
                                        attn_metadata, spec_metadata,
                                        new_tensors_device,
-                                       cache_indirection_buffer)
+                                       cache_indirection_buffer,
+                                       num_accepted_tokens_device,
+                                       req_id_to_old_request)
 
     @torch.inference_mode()
     @with_model_extra_attrs(lambda self: self.model.extra_attrs)
-    def forward(
-        self,
-        scheduled_requests: ScheduledRequests,
-        resource_manager: ResourceManager,
-        new_tensors_device: Optional[SampleStateTensors] = None,
-        gather_context_logits: bool = False,
-        cache_indirection_buffer: Optional[torch.Tensor] = None,
-        spec_decoding_tensor: Optional[SpecDecodingTensor] = None,
-    ):
+    def forward(self,
+                scheduled_requests: ScheduledRequests,
+                resource_manager: ResourceManager,
+                new_tensors_device: Optional[SampleStateTensors] = None,
+                gather_context_logits: bool = False,
+                cache_indirection_buffer: Optional[torch.Tensor] = None,
+                spec_decoding_tensor: Optional[SpecDecodingTensor] = None,
+                num_accepted_tokens_device: Optional[torch.Tensor] = None,
+                req_id_to_old_request: Optional[Dict[int, LlmRequest]] = None):
         kv_cache_manager = resource_manager.get_resource_manager(
             self.kv_cache_manager_key)
 
@@ -2320,11 +2581,20 @@ class PyTorchModelEngine(ModelEngine):
                     return self._forward_step(inputs, gather_ids,
                                               gather_context_logits)
         with self.cuda_graph_runner.pad_batch(
-                scheduled_requests, resource_manager) as padded_requests:
+                scheduled_requests, resource_manager,
+                self.runtime_draft_len) as padded_requests:
 
-            maybe_graph, maybe_attn_metadata, maybe_spec_metadata, key = self.cuda_graph_runner.maybe_get_cuda_graph(
-                padded_requests, spec_resource_manager)
-            if maybe_graph:
+            maybe_attn_metadata, maybe_spec_metadata, key = self.cuda_graph_runner.maybe_get_cuda_graph(
+                padded_requests,
+                enable_spec_decode=self.enable_spec_decode,
+                attn_metadata=attn_metadata,
+                spec_metadata=spec_metadata,
+                draft_tokens_cuda=self.draft_tokens_cuda
+                if self.is_spec_decode else None,
+                spec_resource_manager=spec_resource_manager,
+            )
+            can_run_graph = key is not None
+            if can_run_graph:
                 attn_metadata = maybe_attn_metadata
                 spec_metadata = maybe_spec_metadata
             else:
@@ -2336,11 +2606,11 @@ class PyTorchModelEngine(ModelEngine):
 
             inputs, gather_ids = self._prepare_inputs(
                 padded_requests, kv_cache_manager, attn_metadata, spec_metadata,
-                new_tensors_device, cache_indirection_buffer)
+                new_tensors_device, cache_indirection_buffer,
+                num_accepted_tokens_device, req_id_to_old_request)
 
-            self.iter_counter += 1
             with with_shared_pool(self.cuda_graph_runner.get_graph_pool()):
-                if not maybe_graph:
+                if not can_run_graph:
                     # Fallback to eager execution if graph was not used
                     with MoeLoadBalancerIterContext(moe_load_balancer):
                         outputs = self._forward_step(inputs, gather_ids,
@@ -2358,9 +2628,12 @@ class PyTorchModelEngine(ModelEngine):
                         def capture_postprocess_fn(inputs: Dict[str, Any]):
                             self._postprocess_inputs(inputs)
 
-                        self.cuda_graph_runner.capture(key, capture_forward_fn,
-                                                       inputs,
-                                                       capture_postprocess_fn)
+                        self.cuda_graph_runner.capture(
+                            key,
+                            capture_forward_fn,
+                            inputs,
+                            enable_spec_decode=self.enable_spec_decode,
+                            postprocess_fn=capture_postprocess_fn)
 
                         # here we don't need to use context since cuda graph capture didn't run kernel.
                         # maybe we need a cleaner way to do this.

latest/_modules/index.html

@@ -31,7 +31,7 @@
 <link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../_static/doctools.js?v=9a2dae69"></script>
     <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../genindex.html" />
     <link rel="search" title="Search" href="../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -606,10 +614,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -631,12 +641,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -684,9 +689,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/builder.html

@@ -31,7 +31,7 @@
 <link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../genindex.html" />
     <link rel="search" title="Search" href="../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1834,10 +1842,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1859,12 +1869,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1912,9 +1917,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/disaggregated_params.html

@@ -31,7 +31,7 @@
 <link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../genindex.html" />
     <link rel="search" title="Search" href="../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -637,10 +645,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -662,12 +672,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -715,9 +720,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/executor/request.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -682,10 +690,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -707,12 +717,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -760,9 +765,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/executor/result.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -525,7 +533,7 @@
     <span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm</span><span class="w"> </span><span class="kn">import</span> <span class="n">ray_stub</span> <span class="k">as</span> <span class="n">ray</span>
 
 <span class="kn">from</span><span class="w"> </span><span class="nn">.._ray_utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">unwrap_ray_errors</span>
-<span class="kn">from</span><span class="w"> </span><span class="nn">.._utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">mpi_disabled</span><span class="p">,</span> <span class="n">nvtx_range_debug</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">.._utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">mpi_disabled</span><span class="p">,</span> <span class="n">nvtx_range_debug</span><span class="p">,</span> <span class="n">ray_use_rpc</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">..bindings</span><span class="w"> </span><span class="kn">import</span> <span class="n">executor</span> <span class="k">as</span> <span class="n">tllm</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">..disaggregated_params</span><span class="w"> </span><span class="kn">import</span> <span class="n">DisaggregatedParams</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">..llmapi.tracer</span><span class="w"> </span><span class="kn">import</span> <span class="n">global_tracer</span>
@@ -780,8 +788,10 @@
         <span class="bp">self</span><span class="o">.</span><span class="n">_done</span> <span class="o">=</span> <span class="kc">False</span>
         <span class="bp">self</span><span class="o">.</span><span class="n">metrics_dict</span> <span class="o">=</span> <span class="p">{}</span>
         <span class="bp">self</span><span class="o">.</span><span class="n">trace_headers</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">dict</span><span class="p">[</span><span class="nb">str</span><span class="p">,</span> <span class="nb">str</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+        <span class="c1"># torch backend will use trtllm sampler in beam search mode, but it does not support return logprobs incrementally</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">use_trtllm_sampler</span> <span class="o">=</span> <span class="n">sampling_params</span><span class="o">.</span><span class="n">use_beam_search</span> <span class="ow">and</span> <span class="n">sampling_params</span><span class="o">.</span><span class="n">best_of</span> <span class="o">&gt;</span> <span class="mi">1</span>
 
-        <span class="k">if</span> <span class="n">ray_queue</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+        <span class="k">if</span> <span class="n">ray_queue</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">ray_use_rpc</span><span class="p">():</span>
             <span class="k">if</span> <span class="n">has_event_loop</span><span class="p">():</span>
                 <span class="bp">self</span><span class="o">.</span><span class="n">aqueue</span> <span class="o">=</span> <span class="n">ray_queue</span>
                 <span class="bp">self</span><span class="o">.</span><span class="n">queue</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">aqueue</span>
@@ -886,20 +896,27 @@
             <span class="c1"># each streamed response_tensors.log_probs[src_idx]</span>
             <span class="c1"># contains a streamwise monotonically growing list of logprobs.</span>
             <span class="c1"># so we need to accumulate only the new ones unique to that particular streamed response</span>
-            <span class="k">assert</span> <span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span> <span class="o">&lt;=</span> <span class="nb">len</span><span class="p">(</span>
-                <span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">]</span>
-            <span class="p">),</span> <span class="p">(</span><span class="sa">f</span><span class="s2">&quot;_last_logprobs_len (</span><span class="si">{</span><span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span><span class="si">}</span><span class="s2">) &gt; log_probs length (&quot;</span>
-                <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">])</span><span class="si">}</span><span class="s2">)&quot;</span><span class="p">)</span>
-            <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span> <span class="o">+=</span> <span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">][</span>
-                <span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span><span class="p">:]</span>
+            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_trtllm_sampler</span><span class="p">:</span>
+                <span class="k">assert</span> <span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span> <span class="o">&lt;=</span> <span class="nb">len</span><span class="p">(</span>
+                    <span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">]</span>
+                <span class="p">),</span> <span class="p">(</span><span class="sa">f</span><span class="s2">&quot;_last_logprobs_len (</span><span class="si">{</span><span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span><span class="si">}</span><span class="s2">) &gt; log_probs length (&quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">])</span><span class="si">}</span><span class="s2">)&quot;</span><span class="p">)</span>
+                <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span> <span class="o">+=</span> <span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">][</span>
+                    <span class="n">output</span><span class="o">.</span><span class="n">_last_logprobs_len</span><span class="p">:]</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span> <span class="o">+=</span> <span class="n">response_tensors</span><span class="o">.</span><span class="n">log_probs</span><span class="p">[</span><span class="n">src_idx</span><span class="p">]</span>
+
             <span class="c1"># overcome some WAR in the cpp executor</span>
-            <span class="k">if</span> <span class="n">finish_reasons</span><span class="p">[</span><span class="n">src_idx</span><span class="p">]</span> <span class="o">!=</span> <span class="n">tllm</span><span class="o">.</span><span class="n">FinishReason</span><span class="o">.</span><span class="n">CANCELLED</span><span class="p">:</span>
+            <span class="k">if</span> <span class="n">finish_reasons</span><span class="p">[</span>
+                    <span class="n">src_idx</span><span class="p">]</span> <span class="o">!=</span> <span class="n">tllm</span><span class="o">.</span><span class="n">FinishReason</span><span class="o">.</span><span class="n">CANCELLED</span> <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">use_trtllm_sampler</span><span class="p">:</span>
                 <span class="c1"># Check if logprobs is a list (not a dict or other structure)</span>
                 <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logprobs</span><span class="p">)</span> <span class="o">&gt;</span> <span class="n">output</span><span class="o">.</span><span class="n">length</span><span class="p">:</span>
                     <span class="c1"># LlmResult holds a reference to LogProbStorage, which may be updated by the worker before the result is serialized.</span>
                     <span class="c1"># Therefore, we treat extra logprobs/logits as expected and only consume what&#39;s needed.</span>
                     <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span> <span class="o">=</span> <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span><span class="p">[:</span><span class="n">output</span><span class="o">.</span><span class="n">length</span><span class="p">]</span>
-                <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logprobs</span><span class="p">)</span> <span class="o">==</span> <span class="n">output</span><span class="o">.</span><span class="n">length</span>
+            <span class="k">assert</span> <span class="nb">len</span><span class="p">(</span>
+                <span class="n">output</span><span class="o">.</span><span class="n">logprobs</span>
+            <span class="p">)</span> <span class="o">==</span> <span class="n">output</span><span class="o">.</span><span class="n">length</span><span class="p">,</span> <span class="sa">f</span><span class="s2">&quot;logprobs length: </span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">output</span><span class="o">.</span><span class="n">logprobs</span><span class="p">)</span><span class="si">}</span><span class="s2"> != output.length: </span><span class="si">{</span><span class="n">output</span><span class="o">.</span><span class="n">length</span><span class="si">}</span><span class="s2">&quot;</span>
 
         <span class="k">if</span> <span class="n">response_tensors</span><span class="o">.</span><span class="n">generation_logits</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
             <span class="n">output</span><span class="o">.</span><span class="n">generation_logits</span> <span class="o">=</span> <span class="n">response_tensors</span><span class="o">.</span><span class="n">generation_logits</span><span class="p">[</span>
@@ -1056,7 +1073,7 @@
         <span class="k">else</span><span class="p">:</span>
             <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;Unknown response type: </span><span class="si">{</span><span class="n">response</span><span class="si">}</span><span class="s2">&quot;</span><span class="p">)</span>
 
-        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_done</span> <span class="ow">and</span> <span class="n">mpi_disabled</span><span class="p">():</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_done</span> <span class="ow">and</span> <span class="n">mpi_disabled</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">ray_use_rpc</span><span class="p">():</span>
             <span class="k">assert</span> <span class="nb">hasattr</span><span class="p">(</span>
                 <span class="bp">self</span><span class="o">.</span><span class="n">queue</span><span class="p">,</span> <span class="s2">&quot;unregister&quot;</span>
             <span class="p">),</span> <span class="s2">&quot;Ray path should be activated for unregistering the Ray queue.&quot;</span>
@@ -1289,7 +1306,7 @@
     <span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
         <span class="n">use_async_queue</span> <span class="o">=</span> <span class="n">has_event_loop</span><span class="p">()</span>
         <span class="n">shared_queue</span> <span class="o">=</span> <span class="kc">None</span>
-        <span class="k">if</span> <span class="n">executor</span> <span class="ow">and</span> <span class="n">executor</span><span class="o">.</span><span class="n">use_ray_queue</span><span class="p">():</span>
+        <span class="k">if</span> <span class="n">executor</span> <span class="ow">and</span> <span class="n">executor</span><span class="o">.</span><span class="n">use_ray_queue</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">ray_use_rpc</span><span class="p">():</span>
             <span class="n">shared_queue</span> <span class="o">=</span> <span class="n">executor</span><span class="o">.</span><span class="n">async_response_queue_weakref</span> <span class="k">if</span> <span class="n">use_async_queue</span> <span class="k">else</span> <span class="n">executor</span><span class="o">.</span><span class="n">sync_response_queue_weakref</span>
 
         <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span>
@@ -1354,7 +1371,7 @@
         <span class="k">return</span> <span class="n">response</span>
 
     <span class="k">def</span><span class="w"> </span><span class="nf">_result_step</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">timeout</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">float</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span><span class="p">):</span>
-        <span class="k">if</span> <span class="n">mpi_disabled</span><span class="p">():</span>
+        <span class="k">if</span> <span class="n">mpi_disabled</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">ray_use_rpc</span><span class="p">():</span>
             <span class="k">with</span> <span class="n">unwrap_ray_errors</span><span class="p">():</span>
                 <span class="n">response</span> <span class="o">=</span> <span class="n">ray</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">queue</span><span class="o">.</span><span class="n">get</span><span class="o">.</span><span class="n">remote</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">request_id</span><span class="p">))</span>
             <span class="n">response</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_handle_ray_response</span><span class="p">(</span><span class="n">response</span><span class="p">)</span>
@@ -1365,7 +1382,7 @@
 
     <span class="k">async</span> <span class="k">def</span><span class="w"> </span><span class="nf">_aresult_step</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
         <span class="k">assert</span> <span class="bp">self</span><span class="o">.</span><span class="n">aqueue</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">,</span> <span class="s2">&quot;The asyncio event loop was not present during initialization, so async operations are not available.&quot;</span>
-        <span class="k">if</span> <span class="n">mpi_disabled</span><span class="p">():</span>
+        <span class="k">if</span> <span class="n">mpi_disabled</span><span class="p">()</span> <span class="ow">and</span> <span class="ow">not</span> <span class="n">ray_use_rpc</span><span class="p">():</span>
             <span class="n">response</span> <span class="o">=</span> <span class="k">await</span> <span class="bp">self</span><span class="o">.</span><span class="n">aqueue</span><span class="o">.</span><span class="n">get_async</span><span class="o">.</span><span class="n">remote</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">request_id</span><span class="p">)</span>
             <span class="n">response</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">_handle_ray_response</span><span class="p">(</span><span class="n">response</span><span class="p">)</span>
         <span class="k">else</span><span class="p">:</span>
@@ -1625,10 +1642,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1650,12 +1669,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1703,9 +1717,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/executor/utils.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -713,10 +721,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -738,12 +748,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -791,9 +796,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/functional.html

@@ -31,7 +31,7 @@
 <link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../genindex.html" />
     <link rel="search" title="Search" href="../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -8692,10 +8700,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -8717,12 +8727,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -8770,9 +8775,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/activation.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -564,10 +572,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -589,12 +599,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -642,9 +647,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/attention.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -3427,10 +3435,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -3452,12 +3462,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -3505,9 +3510,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/cast.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -571,10 +579,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -596,12 +606,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -649,9 +654,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/conv.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -820,10 +828,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -845,12 +855,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -898,9 +903,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/embedding.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1287,10 +1295,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1312,12 +1322,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1365,9 +1370,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/linear.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1135,10 +1143,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1160,12 +1170,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1213,9 +1218,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/mlp.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1161,10 +1169,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1186,12 +1196,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1239,9 +1244,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/normalization.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -925,10 +933,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -950,12 +960,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1003,9 +1008,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/layers/pooling.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -580,10 +588,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -605,12 +615,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -658,9 +663,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/llmapi/build_cache.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -863,10 +871,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -888,12 +898,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -941,9 +946,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/llmapi/llm.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1675,10 +1683,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1700,12 +1710,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1753,9 +1758,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/llmapi/llm_args.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -513,9 +521,8 @@
 <span class="kn">from</span><span class="w"> </span><span class="nn">dataclasses</span><span class="w"> </span><span class="kn">import</span> <span class="n">dataclass</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">enum</span><span class="w"> </span><span class="kn">import</span> <span class="n">Enum</span><span class="p">,</span> <span class="n">EnumMeta</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">pathlib</span><span class="w"> </span><span class="kn">import</span> <span class="n">Path</span>
-<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span><span class="n">TYPE_CHECKING</span><span class="p">,</span> <span class="n">Any</span><span class="p">,</span> <span class="n">ClassVar</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span>
-                    <span class="n">Set</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span> <span class="n">Type</span><span class="p">,</span> <span class="n">TypeAlias</span><span class="p">,</span> <span class="n">TypeVar</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">get_args</span><span class="p">,</span>
-                    <span class="n">get_origin</span><span class="p">)</span>
+<span class="kn">from</span><span class="w"> </span><span class="nn">typing</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span><span class="n">Any</span><span class="p">,</span> <span class="n">ClassVar</span><span class="p">,</span> <span class="n">Dict</span><span class="p">,</span> <span class="n">List</span><span class="p">,</span> <span class="n">Literal</span><span class="p">,</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">Set</span><span class="p">,</span> <span class="n">Tuple</span><span class="p">,</span>
+                    <span class="n">Type</span><span class="p">,</span> <span class="n">TypeAlias</span><span class="p">,</span> <span class="n">TypeVar</span><span class="p">,</span> <span class="n">Union</span><span class="p">,</span> <span class="n">get_args</span><span class="p">,</span> <span class="n">get_origin</span><span class="p">)</span>
 
 <span class="kn">import</span><span class="w"> </span><span class="nn">torch</span>
 <span class="kn">import</span><span class="w"> </span><span class="nn">yaml</span>
@@ -530,9 +537,6 @@
 
 <span class="kn">from</span><span class="w"> </span><span class="nn">.._utils</span><span class="w"> </span><span class="kn">import</span> <span class="n">mpi_rank</span>
 
-<span class="k">if</span> <span class="n">TYPE_CHECKING</span><span class="p">:</span>
-    <span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm._torch.pyexecutor.config</span><span class="w"> </span><span class="kn">import</span> <span class="n">PyTorchConfig</span>
-
 <span class="c1"># yapf: disable</span>
 <span class="c1"># isort: off</span>
 <span class="kn">from</span><span class="w"> </span><span class="nn">..bindings.executor</span><span class="w"> </span><span class="kn">import</span> <span class="p">(</span><span class="n">BatchingType</span> <span class="k">as</span> <span class="n">_BatchingType</span><span class="p">,</span>
@@ -730,6 +734,9 @@
 <span class="sd">        &quot;&quot;&quot;</span>
         <span class="k">return</span> <span class="kc">True</span>
 
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_indices_block_size</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="mi">1</span>
+
 
 <div class="viewcode-block" id="RocketSparseAttentionConfig">
 <a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.RocketSparseAttentionConfig">[docs]</a>
@@ -759,6 +766,12 @@
 <a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.RocketSparseAttentionConfig.supports_backend">[docs]</a>
     <span class="k">def</span><span class="w"> </span><span class="nf">supports_backend</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">backend</span><span class="p">:</span> <span class="nb">str</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">bool</span><span class="p">:</span>
         <span class="k">return</span> <span class="n">backend</span> <span class="o">==</span> <span class="s2">&quot;pytorch&quot;</span></div>
+
+
+<div class="viewcode-block" id="RocketSparseAttentionConfig.get_indices_block_size">
+<a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.RocketSparseAttentionConfig.get_indices_block_size">[docs]</a>
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_indices_block_size</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">page_size</span></div>
 </div>
 
 
@@ -794,6 +807,109 @@
 
 
 
+<span class="k">class</span><span class="w"> </span><span class="nc">MoeLoadBalancerConfig</span><span class="p">(</span><span class="n">StrictBaseModel</span><span class="p">):</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">    Pydantic configuration model for the Mixture of Experts (MoE) load balancer.</span>
+
+<span class="sd">    This model holds configuration data (`num_slots`, etc.) as well as</span>
+<span class="sd">    runtime state (`_ep_rank`, `_ep_size`) which must be set via the</span>
+<span class="sd">    `setup()` method before use.</span>
+<span class="sd">    &quot;&quot;&quot;</span>
+
+    <span class="n">num_slots</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
+    <span class="n">initial_global_assignments</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">Dict</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]]</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="nb">repr</span><span class="o">=</span><span class="kc">False</span>  <span class="c1"># Exclude this large dict from model representation</span>
+    <span class="p">)</span>
+    <span class="n">layer_updates_per_iter</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">0</span>
+    <span class="n">_ep_rank</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">PrivateAttr</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+    <span class="n">_ep_size</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">PrivateAttr</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
+
+    <span class="c1"># --- Methods ---</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="nf">setup</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">ep_rank</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">ep_size</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Initializes the runtime state of the configuration.</span>
+<span class="sd">        This must be called before accessing properties like `num_local_slots`.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_ep_rank</span> <span class="o">=</span> <span class="n">ep_rank</span>
+        <span class="bp">self</span><span class="o">.</span><span class="n">_ep_size</span> <span class="o">=</span> <span class="n">ep_size</span>
+
+        <span class="c1"># This assertion was in the original and is critical.</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span><span class="s2">&quot;`num_slots` cannot be None when calling setup().&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span> <span class="o">%</span> <span class="n">ep_size</span> <span class="o">!=</span> <span class="mi">0</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;`num_slots` (</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span><span class="si">}</span><span class="s2">) must be divisible by `ep_size` (</span><span class="si">{</span><span class="n">ep_size</span><span class="si">}</span><span class="s2">).&quot;</span>
+            <span class="p">)</span>
+
+    <span class="c1"># --- Computed Properties ---</span>
+    <span class="c1"># These properties depend on the runtime state set by setup()</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">ep_rank</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Public accessor for the private expert parallel rank.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_rank</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">AttributeError</span><span class="p">(</span><span class="s2">&quot;ep_rank is not set. Call setup() first.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_rank</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">ep_size</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Public accessor for the private expert parallel size.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">AttributeError</span><span class="p">(</span><span class="s2">&quot;ep_size is not set. Call setup() first.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_size</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">num_local_slots</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Calculates the number of slots local to this rank.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_size</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Cannot calculate `num_local_slots`. &quot;</span>
+                <span class="s2">&quot;`num_slots` must be set and setup() must be called.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span> <span class="o">//</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_size</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">slot_start</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Calculates the starting global slot index for this rank.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_rank</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;Cannot calculate `slot_start`. Call setup() first.&quot;</span><span class="p">)</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_ep_rank</span> <span class="o">*</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_local_slots</span>
+
+    <span class="nd">@property</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">slot_end</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="nb">int</span><span class="p">:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Calculates the ending global slot index (exclusive) for this rank.&quot;&quot;&quot;</span>
+        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">slot_start</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_local_slots</span>
+
+    <span class="k">def</span><span class="w"> </span><span class="nf">get_layer_initial_global_assignments</span><span class="p">(</span>
+            <span class="bp">self</span><span class="p">,</span> <span class="n">layer_idx</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]:</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;</span>
+<span class="sd">        Retrieves the initial global assignments for a specific layer.</span>
+<span class="sd">        &quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">initial_global_assignments</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">return</span> <span class="kc">None</span>
+
+        <span class="k">if</span> <span class="n">layer_idx</span> <span class="ow">not</span> <span class="ow">in</span> <span class="bp">self</span><span class="o">.</span><span class="n">initial_global_assignments</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">KeyError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;layer_idx </span><span class="si">{</span><span class="n">layer_idx</span><span class="si">}</span><span class="s2"> not found in `initial_global_assignments`.&quot;</span>
+            <span class="p">)</span>
+
+        <span class="n">assignments</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">initial_global_assignments</span><span class="p">[</span><span class="n">layer_idx</span><span class="p">]</span>
+
+        <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="s2">&quot;`num_slots` is not set, cannot verify assignment length.&quot;</span><span class="p">)</span>
+
+        <span class="k">if</span> <span class="nb">len</span><span class="p">(</span><span class="n">assignments</span><span class="p">)</span> <span class="o">!=</span> <span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span><span class="p">:</span>
+            <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                <span class="sa">f</span><span class="s2">&quot;Assignment length (</span><span class="si">{</span><span class="nb">len</span><span class="p">(</span><span class="n">assignments</span><span class="p">)</span><span class="si">}</span><span class="s2">) for layer </span><span class="si">{</span><span class="n">layer_idx</span><span class="si">}</span><span class="s2"> &quot;</span>
+                <span class="sa">f</span><span class="s2">&quot;does not match `num_slots` (</span><span class="si">{</span><span class="bp">self</span><span class="o">.</span><span class="n">num_slots</span><span class="si">}</span><span class="s2">).&quot;</span><span class="p">)</span>
+
+        <span class="k">return</span> <span class="n">assignments</span>
+
+
 <div class="viewcode-block" id="MoeConfig">
 <a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.MoeConfig">[docs]</a>
 <span class="k">class</span><span class="w"> </span><span class="nc">MoeConfig</span><span class="p">(</span><span class="n">StrictBaseModel</span><span class="p">):</span>
@@ -871,6 +987,7 @@
     <span class="n">moe_tp_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
     <span class="n">moe_ep_size</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="o">-</span><span class="mi">1</span>
     <span class="n">cp_config</span><span class="p">:</span> <span class="nb">dict</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span><span class="n">default_factory</span><span class="o">=</span><span class="nb">dict</span><span class="p">)</span>
+    <span class="n">pp_partition</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
     <span class="n">enable_attention_dp</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
     <span class="n">enable_lm_head_tp_in_adp</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="kc">False</span>
 
@@ -917,6 +1034,7 @@
                        <span class="n">gpus_per_node</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">gpus_per_node</span><span class="p">,</span>
                        <span class="n">tp_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">tp_size</span><span class="p">,</span>
                        <span class="n">pp_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">pp_size</span><span class="p">,</span>
+                       <span class="n">pp_partition</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">pp_partition</span><span class="p">,</span>
                        <span class="n">cp_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cp_size</span><span class="p">,</span>
                        <span class="n">cp_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cp_config</span><span class="p">,</span>
                        <span class="n">enable_attention_dp</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_attention_dp</span><span class="p">,</span>
@@ -1003,6 +1121,16 @@
     <span class="c1"># this value. Otherwise, speculation will always be on.</span>
     <span class="n">max_concurrency</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
 
+    <span class="c1"># Developer interface: dynamically adjust draft length based on active batch size in runtime.</span>
+    <span class="c1"># Maps batch size to draft lengths. For example:</span>
+    <span class="c1"># {1: 4, 4: 2, 8: 0} means:</span>
+    <span class="c1"># - batch_size &gt;= 1: use draft_len=4</span>
+    <span class="c1"># - batch_size &gt;= 4: use draft_len=2</span>
+    <span class="c1"># - batch_size &gt;= 8: use draft_len=0 (disable speculation)</span>
+    <span class="c1"># draft_len_schedule is enforced to contain batch_size=1 and its according draft_len equals max_draft_len for consistency</span>
+    <span class="c1"># for example, if max_draft_len=4, the schedule must contain {1: 4}</span>
+    <span class="n">draft_len_schedule</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">dict</span><span class="p">[</span><span class="nb">int</span><span class="p">,</span> <span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="kc">None</span>
+
     <span class="n">load_format</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">str</span><span class="p">]</span> <span class="o">=</span> <span class="kc">None</span>
     <span class="c1"># PyTorch only.</span>
     <span class="c1"># Rolling average window size (N) for acceptance length across completed requests.</span>
@@ -1040,6 +1168,51 @@
     <span class="c1"># If set, drafting uses greedy sampling, irrespective of sampling parameters.</span>
     <span class="n">_allow_greedy_draft_tokens</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">PrivateAttr</span><span class="p">(</span><span class="kc">True</span><span class="p">)</span>
 
+    <span class="nd">@field_validator</span><span class="p">(</span><span class="s1">&#39;draft_len_schedule&#39;</span><span class="p">)</span>
+    <span class="nd">@classmethod</span>
+    <span class="k">def</span><span class="w"> </span><span class="nf">validate_draft_len_schedule_and_sort</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">v</span><span class="p">,</span> <span class="n">info</span><span class="p">):</span>
+<span class="w">        </span><span class="sd">&quot;&quot;&quot;Validate and sort draft_len_schedule by batch size thresholds.&quot;&quot;&quot;</span>
+        <span class="k">if</span> <span class="n">v</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+            <span class="c1"># Validate values</span>
+            <span class="k">for</span> <span class="n">batch_size</span><span class="p">,</span> <span class="n">draft_len</span> <span class="ow">in</span> <span class="n">v</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                <span class="k">if</span> <span class="n">batch_size</span> <span class="o">&lt;</span> <span class="mi">1</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;draft_len_schedule: batch size threshold must be &gt;= 1, got </span><span class="si">{</span><span class="n">batch_size</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+                <span class="k">if</span> <span class="n">draft_len</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
+                    <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                        <span class="sa">f</span><span class="s2">&quot;draft_len_schedule: draft length must be &gt;= 0, got </span><span class="si">{</span><span class="n">draft_len</span><span class="si">}</span><span class="s2">&quot;</span>
+                    <span class="p">)</span>
+
+            <span class="c1"># Require batch_size=1 in schedule</span>
+            <span class="k">if</span> <span class="mi">1</span> <span class="ow">not</span> <span class="ow">in</span> <span class="n">v</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="s2">&quot;draft_len_schedule must include batch_size=1. &quot;</span>
+                    <span class="s2">&quot;All systems can have batch_size=1. Add {1: &lt;max_draft_len&gt;} to your schedule.&quot;</span>
+                <span class="p">)</span>
+
+            <span class="c1"># Enforce schedule[1] == max_draft_len for consistency</span>
+            <span class="n">max_draft_len</span> <span class="o">=</span> <span class="n">info</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">get</span><span class="p">(</span><span class="s1">&#39;max_draft_len&#39;</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">max_draft_len</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="ow">and</span> <span class="n">v</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span> <span class="o">!=</span> <span class="n">max_draft_len</span><span class="p">:</span>
+                <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                    <span class="sa">f</span><span class="s2">&quot;draft_len_schedule[1] must equal max_draft_len for consistency. &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;Got schedule[1]=</span><span class="si">{</span><span class="n">v</span><span class="p">[</span><span class="mi">1</span><span class="p">]</span><span class="si">}</span><span class="s2">, but max_draft_len=</span><span class="si">{</span><span class="n">max_draft_len</span><span class="si">}</span><span class="s2">. &quot;</span>
+                    <span class="sa">f</span><span class="s2">&quot;batch_size=1 should use maximum draft length.&quot;</span><span class="p">)</span>
+
+            <span class="c1"># Enforce all draft lengths &lt;= max_draft_len</span>
+            <span class="k">if</span> <span class="n">max_draft_len</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
+                <span class="k">for</span> <span class="n">batch_size</span><span class="p">,</span> <span class="n">draft_len</span> <span class="ow">in</span> <span class="n">v</span><span class="o">.</span><span class="n">items</span><span class="p">():</span>
+                    <span class="k">if</span> <span class="n">draft_len</span> <span class="o">&gt;</span> <span class="n">max_draft_len</span><span class="p">:</span>
+                        <span class="k">raise</span> <span class="ne">ValueError</span><span class="p">(</span>
+                            <span class="sa">f</span><span class="s2">&quot;draft_len_schedule: all draft lengths must be &lt;= max_draft_len. &quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot;Got draft_len=</span><span class="si">{</span><span class="n">draft_len</span><span class="si">}</span><span class="s2"> for batch_size=</span><span class="si">{</span><span class="n">batch_size</span><span class="si">}</span><span class="s2">, &quot;</span>
+                            <span class="sa">f</span><span class="s2">&quot;but max_draft_len=</span><span class="si">{</span><span class="n">max_draft_len</span><span class="si">}</span><span class="s2">.&quot;</span><span class="p">)</span>
+
+            <span class="c1"># Return sorted dict (by batch size thresholds)</span>
+            <span class="c1"># This ensures efficient lookup</span>
+            <span class="k">return</span> <span class="nb">dict</span><span class="p">(</span><span class="nb">sorted</span><span class="p">(</span><span class="n">v</span><span class="o">.</span><span class="n">items</span><span class="p">(),</span> <span class="n">key</span><span class="o">=</span><span class="k">lambda</span> <span class="n">x</span><span class="p">:</span> <span class="n">x</span><span class="p">[</span><span class="mi">0</span><span class="p">]))</span>
+        <span class="k">return</span> <span class="n">v</span>
+
     <span class="nd">@classmethod</span>
     <span class="k">def</span><span class="w"> </span><span class="nf">from_dict</span><span class="p">(</span><span class="bp">cls</span><span class="p">,</span> <span class="n">data</span><span class="p">:</span> <span class="nb">dict</span><span class="p">):</span>
         <span class="c1"># dispatch to the correct decoding config</span>
@@ -1445,12 +1618,11 @@
     <span class="c1"># Now we need a flag when MTPDecodingConfig is updated by PyTorchModelEngine.</span>
     <span class="n">num_nextn_predict_layers_from_model_config</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">1</span>
 
-    <span class="c1"># TODO: Hard code for DeepSeek R1</span>
     <span class="c1"># When encounter &lt;think&gt;, start thinking phase.</span>
     <span class="c1"># When encounter &lt;/think&gt;, end thinking phase.</span>
     <span class="c1"># &lt;think&gt; [thinking phase] &lt;/think&gt; [real output]</span>
-    <span class="n">BEGIN_THINKING_PHASE_TOKEN</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">128798</span>
-    <span class="n">END_THINKING_PHASE_TOKEN</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">128799</span>
+    <span class="n">begin_thinking_phase_token</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">128798</span>
+    <span class="n">end_thinking_phase_token</span><span class="p">:</span> <span class="nb">int</span> <span class="o">=</span> <span class="mi">128799</span>
 
 <div class="viewcode-block" id="MTPDecodingConfig.__init__">
 <a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.MTPDecodingConfig.__init__">[docs]</a>
@@ -2158,11 +2330,20 @@
         <span class="s2">&quot;Timeout in milliseconds for KV cache transfer. Requests exceeding this timeout will be cancelled.&quot;</span>
     <span class="p">)</span>
 
+    <span class="n">kv_transfer_sender_future_timeout_ms</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">int</span><span class="p">]</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="mi">1000</span><span class="p">,</span>
+        <span class="n">gt</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span>
+        <span class="n">description</span><span class="o">=</span>
+        <span class="s2">&quot;Timeout in milliseconds to wait for the sender future to be ready when scheduled batch size is 0. This allows the request to be eventually cancelled by the user or because of kv_transfer_timeout_ms&quot;</span>
+    <span class="p">)</span>
+
     <span class="k">def</span><span class="w"> </span><span class="nf">_to_pybind</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
         <span class="k">return</span> <span class="n">_CacheTransceiverConfig</span><span class="p">(</span>
             <span class="n">backend</span><span class="o">=</span><span class="n">_CacheTransceiverBackendType</span><span class="o">.</span><span class="n">from_string</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">backend</span><span class="p">),</span>
             <span class="n">max_tokens_in_buffer</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">max_tokens_in_buffer</span><span class="p">,</span>
-            <span class="n">kv_transfer_timeout_ms</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">kv_transfer_timeout_ms</span><span class="p">)</span></div>
+            <span class="n">kv_transfer_timeout_ms</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">kv_transfer_timeout_ms</span><span class="p">,</span>
+            <span class="n">kv_transfer_sender_future_timeout_ms</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span>
+            <span class="n">kv_transfer_sender_future_timeout_ms</span><span class="p">)</span></div>
 
 
 
@@ -2288,6 +2469,12 @@
         <span class="n">description</span><span class="o">=</span><span class="s2">&quot;Enable LM head TP in attention dp.&quot;</span><span class="p">,</span>
         <span class="n">status</span><span class="o">=</span><span class="s2">&quot;prototype&quot;</span><span class="p">)</span>
 
+    <span class="n">pp_partition</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">List</span><span class="p">[</span><span class="nb">int</span><span class="p">]]</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span>
+        <span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
+        <span class="n">description</span><span class="o">=</span>
+        <span class="s2">&quot;Pipeline parallel partition, a list of each rank&#39;s layer number.&quot;</span><span class="p">,</span>
+        <span class="n">status</span><span class="o">=</span><span class="s2">&quot;prototype&quot;</span><span class="p">)</span>
+
     <span class="n">cp_config</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="nb">dict</span><span class="p">]</span> <span class="o">=</span> <span class="n">Field</span><span class="p">(</span><span class="n">default_factory</span><span class="o">=</span><span class="nb">dict</span><span class="p">,</span>
                                       <span class="n">description</span><span class="o">=</span><span class="s2">&quot;Context parallel config.&quot;</span><span class="p">,</span>
                                       <span class="n">status</span><span class="o">=</span><span class="s2">&quot;prototype&quot;</span><span class="p">)</span>
@@ -2544,6 +2731,7 @@
             <span class="n">moe_ep_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_expert_parallel_size</span><span class="p">,</span>
             <span class="n">enable_attention_dp</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_attention_dp</span><span class="p">,</span>
             <span class="n">enable_lm_head_tp_in_adp</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_lm_head_tp_in_adp</span><span class="p">,</span>
+            <span class="n">pp_partition</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">pp_partition</span><span class="p">,</span>
             <span class="n">cp_config</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cp_config</span><span class="p">)</span>
         <span class="k">return</span> <span class="bp">self</span>
 
@@ -3323,6 +3511,9 @@
     <span class="c1"># PrivateVars</span>
     <span class="n">_quant_config</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">QuantConfig</span><span class="p">]</span> <span class="o">=</span> <span class="n">PrivateAttr</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">None</span><span class="p">)</span>
 
+    <span class="n">_disable_flash_infer_sampling</span><span class="p">:</span> <span class="nb">bool</span> <span class="o">=</span> <span class="n">PrivateAttr</span><span class="p">(</span><span class="n">default</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
+<span class="w">    </span><span class="sd">&quot;&quot;&quot;Unless this is set to False, FlashInfer.sampling is not used, even if available.&quot;&quot;&quot;</span>
+
     <span class="nd">@property</span>
     <span class="k">def</span><span class="w"> </span><span class="nf">quant_config</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="n">QuantConfig</span><span class="p">:</span>
         <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">_quant_config</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
@@ -3405,7 +3596,6 @@
 <a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.TorchLlmArgs.validate_load_balancer">[docs]</a>
     <span class="nd">@model_validator</span><span class="p">(</span><span class="n">mode</span><span class="o">=</span><span class="s2">&quot;after&quot;</span><span class="p">)</span>
     <span class="k">def</span><span class="w"> </span><span class="nf">validate_load_balancer</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="s1">&#39;TorchLlmArgs&#39;</span><span class="p">:</span>
-        <span class="kn">from</span><span class="w"> </span><span class="nn">.._torch</span><span class="w"> </span><span class="kn">import</span> <span class="n">MoeLoadBalancerConfig</span>
         <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">load_balancer</span><span class="p">,</span> <span class="nb">str</span><span class="p">):</span>
             <span class="k">if</span> <span class="ow">not</span> <span class="n">os</span><span class="o">.</span><span class="n">path</span><span class="o">.</span><span class="n">exists</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">load_balancer</span><span class="p">):</span>
                 <span class="k">raise</span> <span class="ne">FileNotFoundError</span><span class="p">(</span>
@@ -3607,82 +3797,6 @@
         <span class="n">executor_config</span> <span class="o">=</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="n">get_executor_config</span><span class="p">(</span><span class="n">_hf_model_dir</span><span class="p">,</span> <span class="n">tokenizer</span><span class="p">)</span>
         <span class="n">executor_config</span><span class="o">.</span><span class="n">mm_encoder_only</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">mm_encoder_only</span>
         <span class="k">return</span> <span class="n">executor_config</span></div>
-
-
-    <span class="c1"># TODO: Remove this after the PyTorch backend is fully migrated to TorchLlmArgs from ExecutorConfig</span>
-<div class="viewcode-block" id="TorchLlmArgs.get_pytorch_backend_config">
-<a class="viewcode-back" href="../../../llm-api/reference.html#tensorrt_llm.llmapi.TorchLlmArgs.get_pytorch_backend_config">[docs]</a>
-    <span class="k">def</span><span class="w"> </span><span class="nf">get_pytorch_backend_config</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="s2">&quot;PyTorchConfig&quot;</span><span class="p">:</span>
-        <span class="kn">from</span><span class="w"> </span><span class="nn">tensorrt_llm._torch.pyexecutor.config</span><span class="w"> </span><span class="kn">import</span> <span class="n">PyTorchConfig</span>
-
-        <span class="k">return</span> <span class="n">PyTorchConfig</span><span class="p">(</span>
-            <span class="n">extra_resource_managers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">extra_resource_managers</span><span class="p">,</span>
-            <span class="n">use_cuda_graph</span><span class="o">=</span><span class="nb">bool</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">),</span>
-            <span class="n">cuda_graph_batch_sizes</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span><span class="o">.</span><span class="n">batch_sizes</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span> <span class="k">else</span>
-            <span class="n">CudaGraphConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;batch_sizes&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">cuda_graph_max_batch_size</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span><span class="o">.</span><span class="n">max_batch_size</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span> <span class="k">else</span>
-            <span class="n">CudaGraphConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;max_batch_size&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">cuda_graph_padding_enabled</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span><span class="o">.</span><span class="n">enable_padding</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">cuda_graph_config</span> <span class="k">else</span>
-            <span class="n">CudaGraphConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;enable_padding&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">disable_overlap_scheduler</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">disable_overlap_scheduler</span><span class="p">,</span>
-            <span class="n">moe_max_num_tokens</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">max_num_tokens</span><span class="p">,</span>
-            <span class="n">moe_load_balancer</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">load_balancer</span><span class="p">,</span>
-            <span class="n">attn_backend</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attn_backend</span><span class="p">,</span>
-            <span class="n">moe_backend</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">backend</span><span class="p">,</span>
-            <span class="n">use_low_precision_moe_combine</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span>
-            <span class="n">use_low_precision_moe_combine</span><span class="p">,</span>
-            <span class="n">sampler_type</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">sampler_type</span><span class="p">,</span>
-            <span class="n">kv_cache_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_config</span><span class="o">.</span><span class="n">dtype</span><span class="p">,</span>
-            <span class="n">mamba_ssm_cache_dtype</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">kv_cache_config</span><span class="o">.</span><span class="n">mamba_ssm_cache_dtype</span><span class="p">,</span>
-            <span class="n">enable_iter_perf_stats</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_iter_perf_stats</span><span class="p">,</span>
-            <span class="n">enable_iter_req_stats</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_iter_req_stats</span><span class="p">,</span>
-            <span class="n">print_iter_log</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">print_iter_log</span><span class="p">,</span>
-            <span class="n">torch_compile_enabled</span><span class="o">=</span><span class="nb">bool</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">),</span>
-            <span class="n">torch_compile_fullgraph</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span><span class="o">.</span><span class="n">enable_fullgraph</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">TorchCompileConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;enable_fullgraph&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">torch_compile_inductor_enabled</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span><span class="o">.</span>
-            <span class="n">enable_inductor</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">TorchCompileConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;enable_inductor&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">torch_compile_piecewise_cuda_graph</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span><span class="o">.</span>
-            <span class="n">enable_piecewise_cuda_graph</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="n">TorchCompileConfig</span><span class="o">.</span>
-            <span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;enable_piecewise_cuda_graph&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">torch_compile_piecewise_cuda_graph_num_tokens</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span>
-            <span class="n">torch_compile_config</span><span class="o">.</span><span class="n">capture_num_tokens</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">TorchCompileConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;capture_num_tokens&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">torch_compile_enable_userbuffers</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span><span class="o">.</span>
-            <span class="n">enable_userbuffers</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">TorchCompileConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;enable_userbuffers&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">torch_compile_max_num_streams</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span><span class="o">.</span>
-            <span class="n">max_num_streams</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">torch_compile_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">TorchCompileConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;max_num_streams&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">enable_autotuner</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_autotuner</span><span class="p">,</span>
-            <span class="n">enable_layerwise_nvtx_marker</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_layerwise_nvtx_marker</span><span class="p">,</span>
-            <span class="n">load_format</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">load_format</span><span class="p">,</span>
-            <span class="n">enable_min_latency</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_min_latency</span><span class="p">,</span>
-            <span class="n">moe_disable_finalize_fusion</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">moe_config</span><span class="o">.</span><span class="n">disable_finalize_fusion</span><span class="p">,</span>
-            <span class="n">stream_interval</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">stream_interval</span><span class="p">,</span>
-            <span class="n">force_dynamic_quantization</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">force_dynamic_quantization</span><span class="p">,</span>
-            <span class="n">allreduce_strategy</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">allreduce_strategy</span><span class="p">,</span>
-            <span class="n">attention_dp_enable_balance</span><span class="o">=</span><span class="nb">bool</span><span class="p">(</span>
-                <span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span>
-                <span class="ow">and</span> <span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span><span class="o">.</span><span class="n">enable_balance</span><span class="p">),</span>
-            <span class="n">attention_dp_time_out_iters</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span><span class="o">.</span><span class="n">timeout_iters</span>
-            <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">AttentionDpConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;timeout_iters&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">attention_dp_batching_wait_iters</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span><span class="o">.</span>
-            <span class="n">batching_wait_iters</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">attention_dp_config</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span>
-            <span class="n">AttentionDpConfig</span><span class="o">.</span><span class="n">model_fields</span><span class="p">[</span><span class="s1">&#39;batching_wait_iters&#39;</span><span class="p">]</span><span class="o">.</span><span class="n">default</span><span class="p">,</span>
-            <span class="n">batch_wait_timeout_ms</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">batch_wait_timeout_ms</span><span class="p">,</span>
-            <span class="n">batch_wait_timeout_iters</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">batch_wait_timeout_iters</span><span class="p">,</span>
-            <span class="n">batch_wait_max_tokens_ratio</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">batch_wait_max_tokens_ratio</span><span class="p">,</span>
-            <span class="n">enable_sleep</span><span class="o">=</span><span class="bp">self</span><span class="o">.</span><span class="n">enable_sleep</span><span class="p">,</span>
-        <span class="p">)</span></div>
 </div>
 
 
@@ -3820,10 +3934,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -3845,12 +3961,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -3898,9 +4009,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/llmapi/mm_encoder.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -690,10 +698,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -715,12 +725,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -768,9 +773,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/llmapi/mpi_session.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1153,10 +1161,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1178,12 +1188,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1231,9 +1236,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/baichuan/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -798,10 +806,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -823,12 +833,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -876,9 +881,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/bert/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1102,10 +1110,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1127,12 +1137,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1180,9 +1185,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/bloom/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -710,10 +718,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -735,12 +745,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -788,9 +793,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/chatglm/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -727,10 +735,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -752,12 +762,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -805,9 +810,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/chatglm/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -926,10 +934,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -951,12 +961,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1004,9 +1009,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/clip/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -755,10 +763,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -780,12 +790,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -833,9 +838,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/cogvlm/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -586,10 +594,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -611,12 +621,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -664,9 +669,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/cogvlm/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -839,10 +847,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -864,12 +874,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -917,9 +922,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/commandr/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -737,10 +745,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -762,12 +772,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -815,9 +820,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/dbrx/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -601,10 +609,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -626,12 +636,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -679,9 +684,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/dbrx/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -727,10 +735,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -752,12 +762,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -805,9 +810,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/deepseek_v1/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -821,10 +829,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -846,12 +856,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -899,9 +904,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/deepseek_v2/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -903,10 +911,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -928,12 +938,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -981,9 +986,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/dit/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -939,10 +947,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -964,12 +974,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1017,9 +1022,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/eagle/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1875,10 +1883,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1900,12 +1910,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1953,9 +1958,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/enc_dec/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -2782,10 +2790,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -2807,12 +2817,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -2860,9 +2865,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/falcon/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -662,10 +670,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -687,12 +697,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -740,9 +745,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/falcon/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -824,10 +832,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -849,12 +859,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -902,9 +907,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gemma/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -752,10 +760,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -777,12 +787,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -830,9 +835,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gemma/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -947,10 +955,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -972,12 +982,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1025,9 +1030,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gpt/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -871,10 +879,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -896,12 +906,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -949,9 +954,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gpt/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -974,10 +982,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -999,12 +1009,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1052,9 +1057,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gptj/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -600,10 +608,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -625,12 +635,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -678,9 +683,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gptj/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -750,10 +758,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -775,12 +785,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -828,9 +833,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/gptneox/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -692,10 +700,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -717,12 +727,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -770,9 +775,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/llama/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -826,10 +834,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -851,12 +861,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -904,9 +909,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/llama/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1174,10 +1182,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1199,12 +1209,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1252,9 +1257,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/mamba/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1019,10 +1027,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1044,12 +1054,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1097,9 +1102,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/medusa/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -659,10 +667,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -684,12 +694,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -737,9 +742,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/medusa/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -809,10 +817,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -834,12 +844,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -887,9 +892,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/mllama/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -2120,10 +2128,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -2145,12 +2155,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -2198,9 +2203,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/mmdit_sd3/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1186,10 +1194,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1211,12 +1221,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1264,9 +1269,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/modeling_utils.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -2595,10 +2603,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -2620,12 +2630,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -2673,9 +2678,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/mpt/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -724,10 +732,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -749,12 +759,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -802,9 +807,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/multimodal_encoders/config.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -658,10 +666,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -683,12 +693,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -736,9 +741,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/multimodal_encoders/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -726,10 +734,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -751,12 +761,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -804,9 +809,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/opt/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -729,10 +737,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -754,12 +764,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -807,9 +812,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/phi/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -771,10 +779,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -796,12 +806,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -849,9 +854,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/phi3/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -867,10 +875,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -892,12 +902,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -945,9 +950,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/recurrentgemma/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1172,10 +1180,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1197,12 +1207,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1250,9 +1255,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/models/redrafter/model.html

@@ -31,7 +31,7 @@
 <link href="../../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../../genindex.html" />
     <link rel="search" title="Search" href="../../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -859,10 +867,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -884,12 +894,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -937,9 +942,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/plugin/plugin.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1316,10 +1324,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1341,12 +1351,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1394,9 +1399,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/quantization/mode.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1017,10 +1025,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1042,12 +1052,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1095,9 +1100,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/quantization/quantize_by_modelopt.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1815,10 +1823,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1840,12 +1850,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1893,9 +1898,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/enc_dec_model_runner.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1086,10 +1094,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1111,12 +1121,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1164,9 +1169,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/generation.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -5426,10 +5434,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -5451,12 +5461,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -5504,9 +5509,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/kv_cache_manager.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1030,10 +1038,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1055,12 +1065,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1108,9 +1113,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/model_runner.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1549,10 +1557,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1574,12 +1584,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1627,9 +1632,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/model_runner_cpp.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1762,10 +1770,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1787,12 +1797,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1840,9 +1845,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/multimodal_model_runner.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -3344,10 +3352,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -3369,12 +3379,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -3422,9 +3427,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/runtime/session.html

@@ -31,7 +31,7 @@
 <link href="../../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../../genindex.html" />
     <link rel="search" title="Search" href="../../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -890,10 +898,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -915,12 +925,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -968,9 +973,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_modules/tensorrt_llm/sampling_params.html

@@ -31,7 +31,7 @@
 <link href="../../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
 
     <link rel="stylesheet" type="text/css" href="../../_static/pygments.css?v=8f2a1f02" />
-    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=df3ac72c" />
+    <link rel="stylesheet" type="text/css" href="../../_static/styles/nvidia-sphinx-theme.css?v=933278ad" />
     <link rel="stylesheet" type="text/css" href="../../_static/copybutton.css?v=76b2166b" />
     <link rel="stylesheet" type="text/css" href="../../_static/autodoc_pydantic.css" />
     <link rel="stylesheet" type="text/css" href="../../_static/togglebutton.css?v=13237357" />
@@ -43,6 +43,8 @@
   <link rel="preload" as="script" href="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
 <link rel="preload" as="script" href="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />
 
+
+
     <script src="../../_static/documentation_options.js?v=5929fcd5"></script>
     <script src="../../_static/doctools.js?v=9a2dae69"></script>
     <script src="../../_static/sphinx_highlight.js?v=dc90522c"></script>
@@ -58,20 +60,24 @@
     <script>
         DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
         DOCUMENTATION_OPTIONS.theme_switcher_json_url = './_static/switcher.json';
-        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc2';
+        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.2.0rc3';
         DOCUMENTATION_OPTIONS.show_version_warning_banner =
             false;
         </script>
+
     <link rel="icon" href="../../_static/favicon.png"/>
+
     <link rel="index" title="Index" href="../../genindex.html" />
     <link rel="search" title="Search" href="../../search.html" />
 
+
   <meta name="viewport" content="width=device-width, initial-scale=1"/>
   <meta name="docsearch:language" content="en"/>
-  <meta name="docsearch:version" content="1.2.0rc2" />
+  <meta name="docsearch:version" content="1.2.0rc3" />
 
 
   </head>
+
   
   
   <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">
@@ -79,6 +85,8 @@
   
   
   <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
+
+
   
   <div id="pst-scroll-pixel-helper"></div>
   
@@ -1085,10 +1093,12 @@
     </div>
   </div>
   
+
   <!-- Scripts loaded after <body> so the DOM is not blocked -->
   <script defer src="../../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
 <script defer src="../../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>
 
+
   <footer class="bd-footer">
 <div class="bd-footer__inner bd-page-width">
   
@@ -1110,12 +1120,7 @@
   
   
   
-  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Manage My Privacy</a>
-   | 
-  
-  
-  
-  <a class="external" href="https://www.nvidia.com/en-us/preferences/start/">Do Not Sell or Share My Data</a>
+  <a class="external" href="https://www.nvidia.com/en-us/about-nvidia/privacy-center/">Your Privacy Choices</a>
    | 
   
   
@@ -1163,9 +1168,9 @@
         <div class="footer-item">
 <div class="extra_footer">
   
-  <p>Last updated on November 05, 2025.</p>
+  <p>Last updated on November 20, 2025.</p>
   
-  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/3111682">3111682</a>.</p>
+  <p>This page is generated by TensorRT-LLM commit <a href="https://github.com/NVIDIA/TensorRT-LLM/tree/2128f73">2128f73</a>.</p>
   
 </div></div>
       

latest/_sources/_cpp_gen/executor.rst.txt

@@ -4,24 +4,6 @@ Executor
 .. Here are files in the cpp/include/executor
 .. We manually add subsection to enable detailed description in the future
 .. It is also doable to automatically generate this file and list all the modules in the conf.py
-transferAgent.h
-_______________
-
-.. doxygenfile:: transferAgent.h
-   :project: TensorRT-LLM
-
-types.h
-_______
-
-.. doxygenfile:: types.h
-   :project: TensorRT-LLM
-
-cacheCommunicator.h
-___________________
-
-.. doxygenfile:: cacheCommunicator.h
-   :project: TensorRT-LLM
-
 disaggServerUtil.h
 __________________
 
@@ -34,6 +16,24 @@ ________
 .. doxygenfile:: tensor.h
    :project: TensorRT-LLM
 
+transferAgent.h
+_______________
+
+.. doxygenfile:: transferAgent.h
+   :project: TensorRT-LLM
+
+serialization.h
+_______________
+
+.. doxygenfile:: serialization.h
+   :project: TensorRT-LLM
+
+types.h
+_______
+
+.. doxygenfile:: types.h
+   :project: TensorRT-LLM
+
 executor.h
 __________
 
@@ -46,9 +46,9 @@ ______________________
 .. doxygenfile:: dataTransceiverState.h
    :project: TensorRT-LLM
 
-serialization.h
-_______________
+cacheCommunicator.h
+___________________
 
-.. doxygenfile:: serialization.h
+.. doxygenfile:: cacheCommunicator.h
    :project: TensorRT-LLM
 

latest/_sources/_cpp_gen/runtime.rst.txt

@@ -4,46 +4,10 @@ Runtime
 .. Here are files in the cpp/include/runtime
 .. We manually add subsection to enable detailed description in the future
 .. It is also doable to automatically generate this file and list all the modules in the conf.py
-iTensor.h
-_________
+lookaheadBuffers.h
+__________________
 
-.. doxygenfile:: iTensor.h
-   :project: TensorRT-LLM
-
-cudaEvent.h
-___________
-
-.. doxygenfile:: cudaEvent.h
-   :project: TensorRT-LLM
-
-virtualMemory.h
-_______________
-
-.. doxygenfile:: virtualMemory.h
-   :project: TensorRT-LLM
-
-speculativeDecodingModule.h
-___________________________
-
-.. doxygenfile:: speculativeDecodingModule.h
-   :project: TensorRT-LLM
-
-common.h
-________
-
-.. doxygenfile:: common.h
-   :project: TensorRT-LLM
-
-samplingConfig.h
-________________
-
-.. doxygenfile:: samplingConfig.h
-   :project: TensorRT-LLM
-
-tllmLogger.h
-____________
-
-.. doxygenfile:: tllmLogger.h
+.. doxygenfile:: lookaheadBuffers.h
    :project: TensorRT-LLM
 
 lookaheadModule.h
@@ -52,88 +16,28 @@ _________________
 .. doxygenfile:: lookaheadModule.h
    :project: TensorRT-LLM
 
+iBuffer.h
+_________
+
+.. doxygenfile:: iBuffer.h
+   :project: TensorRT-LLM
+
 modelConfig.h
 _____________
 
 .. doxygenfile:: modelConfig.h
    :project: TensorRT-LLM
 
-iGptDecoderBatched.h
-____________________
-
-.. doxygenfile:: iGptDecoderBatched.h
-   :project: TensorRT-LLM
-
-cudaStream.h
-____________
-
-.. doxygenfile:: cudaStream.h
-   :project: TensorRT-LLM
-
-loraCache.h
-___________
-
-.. doxygenfile:: loraCache.h
-   :project: TensorRT-LLM
-
-medusaModule.h
-______________
-
-.. doxygenfile:: medusaModule.h
-   :project: TensorRT-LLM
-
-decoderState.h
-______________
-
-.. doxygenfile:: decoderState.h
-   :project: TensorRT-LLM
-
-lookaheadBuffers.h
-__________________
-
-.. doxygenfile:: lookaheadBuffers.h
-   :project: TensorRT-LLM
-
-eagleModule.h
-_____________
-
-.. doxygenfile:: eagleModule.h
-   :project: TensorRT-LLM
-
-runtimeDefaults.h
-_________________
-
-.. doxygenfile:: runtimeDefaults.h
-   :project: TensorRT-LLM
-
 decodingOutput.h
 ________________
 
 .. doxygenfile:: decodingOutput.h
    :project: TensorRT-LLM
 
-decodingInput.h
-_______________
+promptTuningParams.h
+____________________
 
-.. doxygenfile:: decodingInput.h
-   :project: TensorRT-LLM
-
-worldConfig.h
-_____________
-
-.. doxygenfile:: worldConfig.h
-   :project: TensorRT-LLM
-
-gptDecoderBatched.h
-___________________
-
-.. doxygenfile:: gptDecoderBatched.h
-   :project: TensorRT-LLM
-
-explicitDraftTokensBuffers.h
-____________________________
-
-.. doxygenfile:: explicitDraftTokensBuffers.h
+.. doxygenfile:: promptTuningParams.h
    :project: TensorRT-LLM
 
 bufferManager.h
@@ -142,46 +46,22 @@ _______________
 .. doxygenfile:: bufferManager.h
    :project: TensorRT-LLM
 
-loraModule.h
-____________
-
-.. doxygenfile:: loraModule.h
-   :project: TensorRT-LLM
-
-eagleBuffers.h
-______________
-
-.. doxygenfile:: eagleBuffers.h
-   :project: TensorRT-LLM
-
-speculativeDecodingMode.h
-_________________________
-
-.. doxygenfile:: speculativeDecodingMode.h
-   :project: TensorRT-LLM
-
-promptTuningParams.h
-____________________
-
-.. doxygenfile:: promptTuningParams.h
-   :project: TensorRT-LLM
-
-gptDecoder.h
-____________
-
-.. doxygenfile:: gptDecoder.h
-   :project: TensorRT-LLM
-
-memoryCounters.h
-________________
-
-.. doxygenfile:: memoryCounters.h
-   :project: TensorRT-LLM
-
-ipcNvlsMemory.h
+gptJsonConfig.h
 _______________
 
-.. doxygenfile:: ipcNvlsMemory.h
+.. doxygenfile:: gptJsonConfig.h
+   :project: TensorRT-LLM
+
+runtimeDefaults.h
+_________________
+
+.. doxygenfile:: runtimeDefaults.h
+   :project: TensorRT-LLM
+
+loraCache.h
+___________
+
+.. doxygenfile:: loraCache.h
    :project: TensorRT-LLM
 
 rawEngine.h
@@ -190,22 +70,46 @@ ___________
 .. doxygenfile:: rawEngine.h
    :project: TensorRT-LLM
 
-ipcUtils.h
-__________
+gptDecoder.h
+____________
 
-.. doxygenfile:: ipcUtils.h
+.. doxygenfile:: gptDecoder.h
    :project: TensorRT-LLM
 
-iBuffer.h
-_________
+eagleBuffers.h
+______________
 
-.. doxygenfile:: iBuffer.h
+.. doxygenfile:: eagleBuffers.h
    :project: TensorRT-LLM
 
-gptJsonConfig.h
+medusaModule.h
+______________
+
+.. doxygenfile:: medusaModule.h
+   :project: TensorRT-LLM
+
+virtualMemory.h
 _______________
 
-.. doxygenfile:: gptJsonConfig.h
+.. doxygenfile:: virtualMemory.h
+   :project: TensorRT-LLM
+
+explicitDraftTokensBuffers.h
+____________________________
+
+.. doxygenfile:: explicitDraftTokensBuffers.h
+   :project: TensorRT-LLM
+
+iTensor.h
+_________
+
+.. doxygenfile:: iTensor.h
+   :project: TensorRT-LLM
+
+common.h
+________
+
+.. doxygenfile:: common.h
    :project: TensorRT-LLM
 
 loraCachePageManagerConfig.h
@@ -214,3 +118,99 @@ ____________________________
 .. doxygenfile:: loraCachePageManagerConfig.h
    :project: TensorRT-LLM
 
+worldConfig.h
+_____________
+
+.. doxygenfile:: worldConfig.h
+   :project: TensorRT-LLM
+
+loraModule.h
+____________
+
+.. doxygenfile:: loraModule.h
+   :project: TensorRT-LLM
+
+speculativeDecodingMode.h
+_________________________
+
+.. doxygenfile:: speculativeDecodingMode.h
+   :project: TensorRT-LLM
+
+cudaEvent.h
+___________
+
+.. doxygenfile:: cudaEvent.h
+   :project: TensorRT-LLM
+
+decodingInput.h
+_______________
+
+.. doxygenfile:: decodingInput.h
+   :project: TensorRT-LLM
+
+speculativeDecodingModule.h
+___________________________
+
+.. doxygenfile:: speculativeDecodingModule.h
+   :project: TensorRT-LLM
+
+iGptDecoderBatched.h
+____________________
+
+.. doxygenfile:: iGptDecoderBatched.h
+   :project: TensorRT-LLM
+
+eagleModule.h
+_____________
+
+.. doxygenfile:: eagleModule.h
+   :project: TensorRT-LLM
+
+tllmLogger.h
+____________
+
+.. doxygenfile:: tllmLogger.h
+   :project: TensorRT-LLM
+
+gptDecoderBatched.h
+___________________
+
+.. doxygenfile:: gptDecoderBatched.h
+   :project: TensorRT-LLM
+
+cudaStream.h
+____________
+
+.. doxygenfile:: cudaStream.h
+   :project: TensorRT-LLM
+
+ipcNvlsMemory.h
+_______________
+
+.. doxygenfile:: ipcNvlsMemory.h
+   :project: TensorRT-LLM
+
+samplingConfig.h
+________________
+
+.. doxygenfile:: samplingConfig.h
+   :project: TensorRT-LLM
+
+decoderState.h
+______________
+
+.. doxygenfile:: decoderState.h
+   :project: TensorRT-LLM
+
+ipcUtils.h
+__________
+
+.. doxygenfile:: ipcUtils.h
+   :project: TensorRT-LLM
+
+memoryCounters.h
+________________
+
+.. doxygenfile:: memoryCounters.h
+   :project: TensorRT-LLM
+

latest/_sources/commands/trtllm-serve/run-benchmark-with-trtllm-serve.md.txt

@@ -25,7 +25,7 @@ TensorRT LLM distributes the pre-built container on [NGC Catalog](https://catalo
 You can launch the container using the following command:
 
 ```bash
-docker run --rm -it --ipc host -p 8000:8000 --gpus all --ulimit memlock=-1 --ulimit stack=67108864 nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc2
+docker run --rm -it --ipc host -p 8000:8000 --gpus all --ulimit memlock=-1 --ulimit stack=67108864 nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc3
 ```
 
 

latest/_sources/deployment-guide/deployment-guide-for-deepseek-r1-on-trtllm.md.txt

@@ -47,7 +47,7 @@ docker run --rm -it \
 -p 8000:8000 \
 -v ~/.cache:/root/.cache:rw \
 --name tensorrt_llm \
-nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc2 \
+nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc3 \
 /bin/bash
 ```
 

latest/_sources/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md.txt

@@ -43,7 +43,7 @@ docker run --rm -it \
 -p 8000:8000 \
 -v ~/.cache:/root/.cache:rw \
 --name tensorrt_llm \
-nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc2 \
+nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc3 \
 /bin/bash
 ```
 

latest/_sources/deployment-guide/deployment-guide-for-llama3.3-70b-on-trtllm.md.txt

@@ -39,7 +39,7 @@ docker run --rm -it \
 -p 8000:8000 \
 -v ~/.cache:/root/.cache:rw \
 --name tensorrt_llm \
-nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc2 \
+nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc3 \
 /bin/bash
 ```
 

latest/_sources/deployment-guide/deployment-guide-for-llama4-scout-on-trtllm.md.txt

@@ -38,7 +38,7 @@ docker run --rm -it \
 -p 8000:8000 \
 -v ~/.cache:/root/.cache:rw \
 --name tensorrt_llm \
-nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc2 \
+nvcr.io/nvidia/tensorrt-llm/release:1.2.0rc3 \
 /bin/bash
 ```
 

latest/_sources/developer-guide/api-change.md.txt

@@ -34,7 +34,7 @@ TensorRT LLM classifies APIs into two categories:
 All API schemas are:
 - Stored as YAML files in the codebase
 - Protected by unit tests in `tests/unittest/api_stability/`
-- Automatically validated to ensure consistency 
+- Automatically validated to ensure consistency
 
 ## API Change Principles
 
@@ -44,22 +44,26 @@ All API schemas are:
 
 Argument names should describe what the argument represents, not how it is used internally.
 
-✅ **Good**: `max_new_tokens` (clear meaning)  
+✅ **Good**: `max_new_tokens` (clear meaning)
+
 ❌ **Bad**: `num` (ambiguous)
 
 **Reflect Argument Type and Granularity**
 
 - For **boolean** knobs, prefix with verbs like `enable_` and so on.
+
   Examples: `enable_cache`, `enable_flash_attention`
 
-- For **numerical threshold** knobs, suffix with `_limit`, `_size`, `_count`, `_len_` or `_ratio`  
+- For **numerical threshold** knobs, suffix with `_limit`, `_size`, `_count`, `_len_` or `_ratio`
+
   Examples: `max_seq_len`, `prefill_batch_size`
 
 **Avoid Redundant Prefixes**
 
 Example (in `MoeConfig`):
 
-✅ **Good**: `backend`  
+✅ **Good**: `backend`
+
 ❌ **Bad**: `moe_backend` (redundant since it's already in `MoeConfig`)
 
 **Use Specific Names for Narrow Scenarios**
@@ -68,7 +72,8 @@ When adding knobs for specific use cases, make the name convey the restriction c
 
 Example (argument to the LLM class):
 
-✅ **Good**: `rope_scaling_factor` → clearly indicates it's for RoPE  
+✅ **Good**: `rope_scaling_factor` → clearly indicates it's for RoPE
+
 ❌ **Bad**: `scaling_factor` → too generic and prone to misuse
 
 ### 2. Hierarchical Configuration
@@ -77,13 +82,16 @@ Organize complex or hierarchical arguments into **dedicated configuration datacl
 
 **Guidelines**
 
-- Use the `XxxConfig` suffix consistently  
+- Use the `XxxConfig` suffix consistently
+
   Examples: `ModelConfig`, `ParallelConfig`, `MoeConfig`
-  
-- **Reflect conceptual hierarchy**  
+
+- **Reflect conceptual hierarchy**
+
   The dataclass name should represent a coherent functional unit, not an arbitrary grouping
-  
-- **Avoid over-nesting**  
+
+- **Avoid over-nesting**
+
   Use only one level of configuration hierarchy whenever possible (e.g., `LlmArgs → ParallelConfig`) to balance readability and modularity
 
 ### 3. Prefer `LlmArgs` Over Environment Variables
@@ -154,7 +162,7 @@ garbage_collection_gen0_threshold: int = Field(
 
 Add the field to the appropriate schema file:
 
-- **Non-committed arguments**: `tests/unittest/api_stability/references/llm_args.yaml`
+- **Non-committed arguments**: `tests/unittest/api_stability/references/llm.yaml`
   ```yaml
   garbage_collection_gen0_threshold:
     type: int
@@ -162,7 +170,7 @@ Add the field to the appropriate schema file:
     status: beta  # Must match the status in code
   ```
 
-- **Committed arguments**: `tests/unittest/api_stability/references_committed/llm_args.yaml`
+- **Committed arguments**: `tests/unittest/api_stability/references_committed/llm.yaml`
   ```yaml
   garbage_collection_gen0_threshold:
     type: int
@@ -196,16 +204,16 @@ For non-committed APIs, use the `@set_api_status` decorator:
 ```python
 @set_api_status("beta")
 def generate_with_streaming(
-    self, 
-    prompts: List[str], 
+    self,
+    prompts: List[str],
     **kwargs
 ) -> Iterator[GenerationOutput]:
     """Generate text with streaming output.
-    
+
     Args:
         prompts: Input prompts for generation
         **kwargs: Additional generation parameters
-        
+
     Returns:
         Iterator of generation outputs
     """

latest/_sources/examples/curl_chat_client.rst.txt

@@ -2,7 +2,7 @@ Curl Chat Client
 ================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/curl_chat_client.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/curl_chat_client.sh.
 
 .. literalinclude:: ../../../examples/serve/curl_chat_client.sh
     :lines: 1-11

latest/_sources/examples/curl_chat_client_for_multimodal.rst.txt

@@ -2,7 +2,7 @@ Curl Chat Client For Multimodal
 ===============================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/curl_chat_client_for_multimodal.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/curl_chat_client_for_multimodal.sh.
 
 .. literalinclude:: ../../../examples/serve/curl_chat_client_for_multimodal.sh
     :lines: 1-88

latest/_sources/examples/curl_completion_client.rst.txt

@@ -2,7 +2,7 @@ Curl Completion Client
 ======================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/curl_completion_client.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/curl_completion_client.sh.
 
 .. literalinclude:: ../../../examples/serve/curl_completion_client.sh
     :lines: 1-10

latest/_sources/examples/deepseek_r1_reasoning_parser.rst.txt

@@ -2,7 +2,7 @@ Deepseek R1 Reasoning Parser
 ============================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/deepseek_r1_reasoning_parser.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/deepseek_r1_reasoning_parser.sh.
 
 .. literalinclude:: ../../../examples/serve/deepseek_r1_reasoning_parser.sh
     :lines: 1-10

latest/_sources/examples/genai_perf_client.rst.txt

@@ -2,7 +2,7 @@ Genai Perf Client
 =================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/genai_perf_client.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/genai_perf_client.sh.
 
 .. literalinclude:: ../../../examples/serve/genai_perf_client.sh
     :lines: 1-16

latest/_sources/examples/genai_perf_client_for_multimodal.rst.txt

@@ -2,7 +2,7 @@ Genai Perf Client For Multimodal
 ================================
 Refer to the `trtllm-serve documentation <https://nvidia.github.io/TensorRT-LLM/commands/trtllm-serve.html>`_ for starting a server.
 
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/serve/genai_perf_client_for_multimodal.sh.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/serve/genai_perf_client_for_multimodal.sh.
 
 .. literalinclude:: ../../../examples/serve/genai_perf_client_for_multimodal.sh
     :lines: 1-19

latest/_sources/examples/llm_guided_decoding.rst.txt

@@ -1,6 +1,6 @@
 Generate text with guided decoding
 ==================================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_guided_decoding.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_guided_decoding.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_guided_decoding.py
     :lines: 4-47

latest/_sources/examples/llm_inference.rst.txt

@@ -1,6 +1,6 @@
 Generate text
 =============
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_inference.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_inference.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_inference.py
     :lines: 4-35

latest/_sources/examples/llm_inference_async.rst.txt

@@ -1,6 +1,6 @@
 Generate text asynchronously
 ============================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_inference_async.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_inference_async.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_inference_async.py
     :lines: 4-43

latest/_sources/examples/llm_inference_async_streaming.rst.txt

@@ -1,6 +1,6 @@
 Generate text in streaming
 ==========================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_inference_async_streaming.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_inference_async_streaming.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_inference_async_streaming.py
     :lines: 4-64

latest/_sources/examples/llm_inference_distributed.rst.txt

@@ -1,6 +1,6 @@
 Distributed LLM Generation
 ==========================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_inference_distributed.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_inference_distributed.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_inference_distributed.py
     :lines: 4-44

latest/_sources/examples/llm_kv_cache_connector.rst.txt

@@ -1,6 +1,6 @@
 KV Cache Connector
 ==================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_kv_cache_connector.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_kv_cache_connector.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_kv_cache_connector.py
     :lines: 4-247

latest/_sources/examples/llm_kv_cache_offloading.rst.txt

@@ -1,6 +1,6 @@
 KV Cache Offloading
 ===================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_kv_cache_offloading.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_kv_cache_offloading.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_kv_cache_offloading.py
     :lines: 4-134

latest/_sources/examples/llm_logits_processor.rst.txt

@@ -1,6 +1,6 @@
 Control generated text using logits processor
 =============================================
-Source https://github.com/NVIDIA/TensorRT-LLM/blob/31116825b39f4e6a6a1e127001f5204b73d1dc32/examples/llm-api/llm_logits_processor.py.
+Source https://github.com/NVIDIA/TensorRT-LLM/blob/2128f73d58508a1a0b37119bd851edb19ab88635/examples/llm-api/llm_logits_processor.py.
 
 .. literalinclude:: ../../../examples/llm-api/llm_logits_processor.py
     :lines: 4-128