import inspect import json import pickle import tempfile from pathlib import Path from typing import List import numpy as np import torch import tensorrt_llm.bindings as _tb def test_generation_output(): ids = torch.ones(1) lengths = torch.ones(2) gen_output = _tb.GenerationOutput(ids, lengths) assert torch.equal(gen_output.ids, ids) assert torch.equal(gen_output.lengths, lengths) assert gen_output.log_probs is None log_probs = torch.ones(1) gen_output.log_probs = log_probs assert gen_output.log_probs == log_probs assert gen_output.context_logits is None torch.ones(1) gen_output.context_logits = log_probs assert gen_output.context_logits == log_probs def test_generation_input(): end_id = 42 pad_id = 13 ids = torch.ones(1) lengths = torch.ones(2) packed = True gen_input = _tb.GenerationInput(end_id, pad_id, ids, lengths, packed) assert gen_input.end_id == end_id assert gen_input.pad_id == pad_id assert torch.equal(gen_input.ids, ids) assert torch.equal(gen_input.lengths, lengths) assert gen_input.packed == packed assert gen_input.max_new_tokens is None max_new_tokens = 100 gen_input.max_new_tokens = max_new_tokens assert gen_input.max_new_tokens == max_new_tokens assert gen_input.embedding_bias is None embedding_bias = torch.ones(3) gen_input.embedding_bias = embedding_bias assert torch.equal(gen_input.embedding_bias, embedding_bias) assert gen_input.prompt_tuning_params.embedding_table is None assert gen_input.prompt_tuning_params.tasks is None assert gen_input.prompt_tuning_params.vocab_size is None embedding_table = torch.ones(3) tasks = torch.ones(2) vocab_size = torch.ones(1) prompt_tuning_params = _tb.PromptTuningParams( embedding_table=embedding_table, tasks=tasks, vocab_size=vocab_size) assert len(prompt_tuning_params.prompt_tuning_enabled) == 0 prompt_tuning_enabled = [True, False] prompt_tuning_params.prompt_tuning_enabled = prompt_tuning_enabled assert len(prompt_tuning_params.prompt_tuning_enabled) == 2 assert prompt_tuning_params.prompt_tuning_enabled == prompt_tuning_enabled gen_input.prompt_tuning_params = prompt_tuning_params assert gen_input.prompt_tuning_params is not None assert torch.equal(gen_input.prompt_tuning_params.embedding_table, embedding_table) assert torch.equal(gen_input.prompt_tuning_params.tasks, tasks) assert torch.equal(gen_input.prompt_tuning_params.vocab_size, vocab_size) assert gen_input.prompt_tuning_params.prompt_tuning_enabled == prompt_tuning_enabled def test_gpt_session_config(): kv_cache_config = _tb.KvCacheConfig() assert kv_cache_config.max_tokens is None max_tokens = 13 kv_cache_config.max_tokens = max_tokens assert kv_cache_config.max_tokens == max_tokens assert kv_cache_config.free_gpu_memory_fraction is None free_gpu_memory_fraction = 0.5 kv_cache_config.free_gpu_memory_fraction = free_gpu_memory_fraction assert kv_cache_config.free_gpu_memory_fraction == free_gpu_memory_fraction max_batch_size = 1000 max_beam_width = 64 max_sequence_length = 1 << 20 gpt_session_config = _tb.GptSessionConfig(max_batch_size, max_beam_width, max_sequence_length) assert gpt_session_config.max_batch_size == max_batch_size assert gpt_session_config.max_beam_width == max_beam_width assert gpt_session_config.max_sequence_length == max_sequence_length assert gpt_session_config.kv_cache_config is not None assert gpt_session_config.kv_cache_config.max_tokens is None assert gpt_session_config.kv_cache_config.free_gpu_memory_fraction is None gpt_session_config.kv_cache_config = kv_cache_config assert gpt_session_config.kv_cache_config.max_tokens == max_tokens assert gpt_session_config.kv_cache_config.free_gpu_memory_fraction == free_gpu_memory_fraction gpt_session_config.kv_cache_config.max_tokens = None assert gpt_session_config.kv_cache_config.max_tokens is None gpt_session_config.kv_cache_config.free_gpu_memory_fraction = None assert gpt_session_config.kv_cache_config.free_gpu_memory_fraction is None assert not gpt_session_config.decoder_per_request gpt_session_config.decoder_per_request = True assert gpt_session_config.decoder_per_request assert not gpt_session_config.cuda_graph_mode gpt_session_config.cuda_graph_mode = True assert gpt_session_config.cuda_graph_mode assert gpt_session_config.ctx_micro_batch_size is None ctx_micro_batch_size = 10 gpt_session_config.ctx_micro_batch_size = ctx_micro_batch_size assert gpt_session_config.ctx_micro_batch_size == ctx_micro_batch_size assert gpt_session_config.gen_micro_batch_size is None gen_micro_batch_size = 20 gpt_session_config.gen_micro_batch_size = gen_micro_batch_size assert gpt_session_config.gen_micro_batch_size == gen_micro_batch_size def test_quant_mode(): assert _tb.QuantMode.none().value == 0 assert _tb.QuantMode.int4_weights().has_int4_weights assert _tb.QuantMode.int8_weights().has_int8_weights assert _tb.QuantMode.activations().has_activations assert _tb.QuantMode.per_channel_scaling().has_per_channel_scaling assert _tb.QuantMode.per_token_scaling().has_per_token_scaling assert _tb.QuantMode.per_group_scaling().has_per_group_scaling assert _tb.QuantMode.int8_kv_cache().has_int8_kv_cache assert _tb.QuantMode.fp8_kv_cache().has_fp8_kv_cache assert _tb.QuantMode.fp8_qdq().has_fp8_qdq quant_mode = _tb.QuantMode.from_description(True, True, True, True, True, True, True, True) assert quant_mode.has_int4_weights quant_mode -= _tb.QuantMode.int4_weights() assert not quant_mode.has_int4_weights quant_mode += _tb.QuantMode.int4_weights() assert quant_mode.has_int4_weights assert _tb.QuantMode.none() == _tb.QuantMode.none() def test_decoding_mode(): assert _tb.DecodingMode.none().is_none assert not _tb.DecodingMode.none().is_top_k assert not _tb.DecodingMode.none().is_top_p assert not _tb.DecodingMode.none().is_beam_search assert _tb.DecodingMode.top_k().is_top_k assert _tb.DecodingMode.top_k().is_top_k_or_top_p assert not _tb.DecodingMode.top_k().is_top_p assert not _tb.DecodingMode.top_k().is_top_k_and_top_p assert not _tb.DecodingMode.top_k().is_none assert not _tb.DecodingMode.top_k().is_beam_search assert _tb.DecodingMode.top_p().is_top_p assert _tb.DecodingMode.top_p().is_top_k_or_top_p assert not _tb.DecodingMode.top_p().is_top_k assert not _tb.DecodingMode.top_p().is_top_k_and_top_p assert not _tb.DecodingMode.top_p().is_none assert not _tb.DecodingMode.top_p().is_beam_search assert _tb.DecodingMode.top_k_top_p().is_top_p assert _tb.DecodingMode.top_k_top_p().is_top_k assert _tb.DecodingMode.top_k_top_p().is_top_k_or_top_p assert _tb.DecodingMode.top_k_top_p().is_top_k_and_top_p assert not _tb.DecodingMode.top_k_top_p().is_none assert not _tb.DecodingMode.top_k_top_p().is_beam_search assert _tb.DecodingMode.beam_search().is_beam_search assert not _tb.DecodingMode.beam_search().is_none assert not _tb.DecodingMode.beam_search().is_top_k assert not _tb.DecodingMode.beam_search().is_top_p def test_gpt_model_config(): vocab_size = 10000 num_layers = 12 num_heads = 16 hidden_size = 768 data_type = _tb.DataType.FLOAT gpt_model_config = _tb.GptModelConfig(vocab_size, num_layers, num_heads, hidden_size, data_type) assert gpt_model_config.vocab_size == vocab_size assert gpt_model_config.num_layers() == num_layers assert gpt_model_config.num_heads == num_heads assert gpt_model_config.hidden_size == hidden_size assert gpt_model_config.data_type == data_type assert gpt_model_config.vocab_size_padded(1) is not None assert gpt_model_config.size_per_head == hidden_size // num_heads assert gpt_model_config.num_kv_heads == num_heads num_kv_heads = 1 gpt_model_config.num_kv_heads = num_kv_heads assert gpt_model_config.num_kv_heads == num_kv_heads assert not gpt_model_config.use_gpt_attention_plugin gpt_model_config.use_gpt_attention_plugin = True assert gpt_model_config.use_gpt_attention_plugin assert not gpt_model_config.use_packed_input gpt_model_config.use_packed_input = True assert gpt_model_config.use_packed_input assert not gpt_model_config.use_paged_kv_cache gpt_model_config.use_paged_kv_cache = True assert gpt_model_config.use_paged_kv_cache assert gpt_model_config.tokens_per_block == 64 tokens_per_block = 1024 gpt_model_config.tokens_per_block = tokens_per_block assert gpt_model_config.tokens_per_block == tokens_per_block assert gpt_model_config.quant_mode == _tb.QuantMode.none() gpt_model_config.quant_mode = _tb.QuantMode.int4_weights() assert gpt_model_config.quant_mode.has_int4_weights assert gpt_model_config.supports_inflight_batching assert gpt_model_config.max_batch_size == 0 max_batch_size = 1000 gpt_model_config.max_batch_size = max_batch_size assert gpt_model_config.max_batch_size == max_batch_size assert gpt_model_config.max_input_len == 0 max_input_len = 2048 gpt_model_config.max_input_len = max_input_len assert gpt_model_config.max_input_len == max_input_len assert gpt_model_config.max_num_tokens is None max_num_tokens = 10000 gpt_model_config.max_num_tokens = max_num_tokens assert gpt_model_config.max_num_tokens == max_num_tokens assert not gpt_model_config.compute_context_logits gpt_model_config.compute_context_logits = True assert gpt_model_config.compute_context_logits assert not gpt_model_config.compute_generation_logits gpt_model_config.compute_generation_logits = True assert gpt_model_config.compute_generation_logits assert gpt_model_config.model_variant == _tb.GptModelVariant.GPT model_variant = _tb.GptModelVariant.GLM gpt_model_config.model_variant = model_variant assert gpt_model_config.model_variant == model_variant assert not gpt_model_config.use_custom_all_reduce gpt_model_config.use_custom_all_reduce = True assert gpt_model_config.use_custom_all_reduce def test_world_config(): tensor_parallelism = 2 pipeline_parallelism = 4 rank = 3 gpus_per_node = 10 world_config = _tb.WorldConfig(tensor_parallelism, pipeline_parallelism, rank, gpus_per_node) assert world_config.tensor_parallelism == tensor_parallelism assert world_config.pipeline_parallelism == pipeline_parallelism assert world_config.rank == rank assert world_config.gpus_per_node == gpus_per_node assert world_config.gpus_per_group == gpus_per_node assert world_config.size == tensor_parallelism * pipeline_parallelism assert world_config.is_pipeline_parallel assert world_config.is_tensor_parallel assert world_config.device == rank % gpus_per_node assert world_config.pipeline_parallel_rank == rank // tensor_parallelism assert world_config.tensor_parallel_rank == rank % tensor_parallelism world_config = _tb.WorldConfig.mpi(gpus_per_node) assert world_config.tensor_parallelism == 1 assert world_config.pipeline_parallelism == 1 assert world_config.gpus_per_node == gpus_per_node assert world_config.rank == 0 gpus_per_group = gpus_per_node // 2 device_ids = list(gpus_per_group + x for x in range(gpus_per_group)) assert max(device_ids) < gpus_per_node world_config = _tb.WorldConfig(rank=rank, gpus_per_node=gpus_per_node, device_ids=device_ids) assert world_config.gpus_per_node == gpus_per_node assert world_config.gpus_per_group == gpus_per_group assert world_config.rank == rank assert world_config.device == rank + gpus_per_group def test_sampling_config(): beam_width = 12 sampling_config = _tb.SamplingConfig(beam_width) assert sampling_config.beam_width == 12 def check_empty_then_set(member, value): assert getattr(sampling_config, member) is None setattr(sampling_config, member, value) assert getattr(sampling_config, member) == value float_array = [1., 2., 3.] size_t_array = [1, 2, 3] check_empty_then_set("temperature", float_array) check_empty_then_set("min_length", size_t_array) check_empty_then_set("repetition_penalty", float_array) check_empty_then_set("presence_penalty", float_array) check_empty_then_set("frequency_penalty", float_array) check_empty_then_set("top_k", size_t_array) check_empty_then_set("top_p", float_array) check_empty_then_set("random_seed", size_t_array) check_empty_then_set("top_p_decay", float_array) check_empty_then_set("top_p_min", float_array) check_empty_then_set("top_p_reset_ids", size_t_array) check_empty_then_set("beam_search_diversity_rate", float_array) check_empty_then_set("length_penalty", float_array) check_empty_then_set("early_stopping", size_t_array) def test_gpt_json_config(): model_config = { "vocab_size": 1000, "num_layers": 12, "num_heads": 4, "hidden_size": 512, "data_type": _tb.DataType.FLOAT, } gpt_model_config = _tb.GptModelConfig(**model_config) json_config = { "name": "gpt", "version": "none", "precision": "float32", "tensor_parallelism": 1, "pipeline_parallelism": 1, "model_config": gpt_model_config } gpt_json_config = _tb.GptJsonConfig(**json_config) def check_properties(the_object, properties, model_config): for property, value in properties.items(): if isinstance(value, _tb.GptModelConfig): object_config = getattr(the_object, property) for subproperty, subvalue in model_config.items(): member = getattr(object_config, subproperty) if callable(member): member = member() assert member == subvalue else: assert getattr(the_object, property) == value check_properties(gpt_json_config, json_config, model_config) json_dict = { "builder_config": { "name": json_config["name"], "vocab_size": model_config["vocab_size"], "num_layers": model_config["num_layers"], "num_heads": model_config["num_heads"], "hidden_size": model_config["hidden_size"], "precision": json_config["precision"], "tensor_parallel": json_config["tensor_parallelism"], "pipeline_parallel": json_config["pipeline_parallelism"], }, "plugin_config": { "paged_kv_cache": False, "tokens_per_block": 0, "gpt_attention_plugin": False, "remove_input_padding": False, "use_custom_all_reduce": False, "use_context_fmha_for_generation": False, "use_paged_context_fmha": False, "lora_plugin": False, } } gpt_json_config = _tb.GptJsonConfig.parse(json.dumps(json_dict)) with tempfile.NamedTemporaryFile("w", delete=False) as fp: json.dump(json_dict, fp) fp.close() gpt_json_config = _tb.GptJsonConfig.parse_file(Path(fp.name)) Path(fp.name).unlink() rank = 3 gpus_per_node = 10 world_config = _tb.WorldConfig(json_config["tensor_parallelism"], json_config["pipeline_parallelism"], rank, gpus_per_node) assert gpt_json_config.engine_filename( world_config) == json_config["name"] + "_float32_tp1_rank3.engine" assert gpt_json_config.engine_filename( world_config, "llama") == "llama_float32_tp1_rank3.engine" def test_gpt_session(): members = {name: tpe for (name, tpe) in inspect.getmembers(_tb.GptSession)} assert isinstance(members["model_config"], property) assert isinstance(members["world_config"], property) assert isinstance(members["device"], property) assert "generate" in members def test_llm_request(): beam_width = 2 sampling_config = _tb.SamplingConfig(beam_width) kwargs = { "request_id": 0, "max_new_tokens": 5, "sampling_config": sampling_config, "input_tokens": [0, 1, 2], "is_streaming": True, "pad_id": 99, "end_id": 100, "prompt_embedding_table": torch.tensor((10, 10)), "prompt_vocab_size": 2, "embedding_bias": torch.tensor((10, 10)), "stop_words_list": torch.tensor((10, 10)), "bad_words_list": torch.tensor((10, 10)), "return_log_probs": True, "return_context_logits": False, "return_generation_logits": False } llm_request = _tb.LlmRequest(**kwargs) assert llm_request.request_id == 0 assert llm_request.prompt_len == 3 assert llm_request.sampling_config.beam_width == sampling_config.beam_width assert llm_request.is_streaming assert llm_request.pad_id == 99 assert llm_request.end_id == 100 assert llm_request.seq_slot == None assert torch.equal(llm_request.prompt_embedding_table, kwargs["prompt_embedding_table"]) assert llm_request.prompt_vocab_size == 2 assert torch.equal(llm_request.embedding_bias, kwargs["embedding_bias"]) assert torch.equal(llm_request.stop_words_list, kwargs["stop_words_list"]) assert torch.equal(llm_request.bad_words_list, kwargs["bad_words_list"]) assert llm_request.get_num_tokens(0) == 3 assert llm_request.max_beam_num_tokens == 3 assert llm_request.get_token(1, 2) == 2 assert llm_request.get_tokens(1) == [0, 1, 2] assert llm_request.max_num_generated_tokens == 0 llm_request.add_new_token(42, 0) assert llm_request.get_token(0, 3) == 42 llm_request.add_new_tokens([43, 44]) assert llm_request.get_token(0, 4) == 43 assert llm_request.get_token(1, 3) == 44 llm_request.set_generated_tokens([[10, 11], [12, 13]]) assert llm_request.get_tokens(0) == [0, 1, 2, 10, 11] assert llm_request.max_num_generated_tokens == 2 llm_request.pause(0) assert llm_request.state == _tb.LlmRequestState.REQUEST_STATE_CONTEXT_INIT llm_request.max_sent_token_pos = 1 assert llm_request.max_sent_token_pos == 1 assert llm_request.return_log_probs llm_request.set_log_probs([0.1], 0) llm_request.set_log_probs([0.2], 1) assert np.allclose(llm_request.get_log_probs(0), np.array([0.1])) assert np.allclose(llm_request.log_probs, np.array([[0.1], [0.2]])) llm_request.set_cum_log_prob(0.1, 0) llm_request.set_cum_log_prob(0.2, 1) assert np.allclose(llm_request.cum_log_probs, np.array([0.1, 0.2])) assert llm_request.orig_prompt_len == 3 assert not llm_request.draft_tokens llm_request.draft_tokens = [1, 2, 3] assert llm_request.draft_tokens == [1, 2, 3] logits = torch.tensor([-5, -6 - 7], dtype=torch.float) llm_request.draft_logits = logits assert torch.equal(llm_request.draft_logits, logits) def test_inference_request(): input_ids = torch.tensor((10, 10)) def logits_post_processor(req_id: int, logits: torch.Tensor, ids: List[List[int]]): del req_id, ids ir = _tb.InferenceRequest(42, logits_post_processor) setattr(ir, _tb.tensor_names.INPUT_IDS, input_ids) assert ir.request_id == 42 assert ir.input_ids is not None assert torch.equal(ir.input_ids, input_ids) assert not ir.is_streaming ir.is_streaming = True assert ir.is_streaming data_tensor = torch.tensor((5, 5)) assert ir.draft_input_ids is None ir.draft_input_ids = data_tensor assert torch.equal(ir.draft_input_ids, data_tensor) assert ir.draft_logits is None ir.draft_logits = data_tensor assert torch.equal(ir.draft_logits, data_tensor) assert ir.bad_words_list is None ir.bad_words_list = data_tensor assert torch.equal(ir.bad_words_list, data_tensor) assert ir.beam_width is None ir.beam_width = data_tensor assert torch.equal(ir.beam_width, data_tensor) assert ir.embedding_bias is None ir.embedding_bias = data_tensor assert torch.equal(ir.embedding_bias, data_tensor) assert ir.end_id is None ir.end_id = data_tensor assert torch.equal(ir.end_id, data_tensor) assert ir.length_penalty is None ir.length_penalty = data_tensor assert torch.equal(ir.length_penalty, data_tensor) assert ir.early_stopping is None ir.early_stopping = data_tensor assert torch.equal(ir.early_stopping, data_tensor) assert ir.max_new_tokens is None ir.max_new_tokens = data_tensor assert torch.equal(ir.max_new_tokens, data_tensor) assert ir.min_length is None ir.min_length = data_tensor assert torch.equal(ir.min_length, data_tensor) assert ir.pad_id is None ir.pad_id = data_tensor assert torch.equal(ir.pad_id, data_tensor) assert ir.presence_penalty is None ir.presence_penalty = data_tensor assert torch.equal(ir.presence_penalty, data_tensor) assert ir.frequency_penalty is None ir.frequency_penalty = data_tensor assert torch.equal(ir.frequency_penalty, data_tensor) assert ir.prompt_embedding_table is None ir.prompt_embedding_table = data_tensor assert torch.equal(ir.prompt_embedding_table, data_tensor) assert ir.prompt_vocab_size is None ir.prompt_vocab_size = data_tensor assert torch.equal(ir.prompt_vocab_size, data_tensor) assert ir.lora_weights is None ir.lora_weights = data_tensor assert torch.equal(ir.lora_weights, data_tensor) assert ir.lora_config is None ir.lora_config = data_tensor assert torch.equal(ir.lora_config, data_tensor) assert ir.random_seed is None ir.random_seed = data_tensor assert torch.equal(ir.random_seed, data_tensor) assert ir.repetition_penalty is None ir.repetition_penalty = data_tensor assert torch.equal(ir.repetition_penalty, data_tensor) assert ir.return_log_probs is None ir.return_log_probs = data_tensor assert torch.equal(ir.return_log_probs, data_tensor) assert ir.runtime_top_k is None ir.runtime_top_k = data_tensor assert torch.equal(ir.runtime_top_k, data_tensor) assert ir.runtime_top_p is None ir.runtime_top_p = data_tensor assert torch.equal(ir.runtime_top_p, data_tensor) assert ir.stop_words_list is None ir.stop_words_list = data_tensor assert torch.equal(ir.stop_words_list, data_tensor) assert ir.temperature is None ir.temperature = data_tensor assert torch.equal(ir.temperature, data_tensor) ir.logits_post_processor = None serialized = pickle.dumps(ir) deserialized = pickle.loads(serialized) assert isinstance(deserialized, _tb.InferenceRequest) assert deserialized.request_id == ir.request_id assert deserialized.is_streaming == ir.is_streaming assert torch.equal(deserialized.input_ids, ir.input_ids) def test_trt_gpt_model_optional_params(): opt_params = _tb.TrtGptModelOptionalParams() kv_cache_config = _tb.KvCacheConfig(10, 10, 0, 0.5, False) opt_params.kv_cache_config = kv_cache_config assert opt_params.kv_cache_config.free_gpu_memory_fraction == kv_cache_config.free_gpu_memory_fraction opt_params.enable_trt_overlap = True assert opt_params.enable_trt_overlap assert opt_params.device_ids is None opt_params.device_ids = [0, 1] assert opt_params.device_ids == [0, 1] def test_KvCacheConfig_pickle(): cache = _tb.KvCacheConfig(free_gpu_memory_fraction=0.4) cache1 = pickle.dumps(cache) cache2 = pickle.loads(cache1) assert cache2 == cache def test_TrtGptModelOptionalParams_pickle(): cache = _tb.KvCacheConfig(free_gpu_memory_fraction=0.4) params1 = _tb.TrtGptModelOptionalParams( kv_cache_config=cache, enable_trt_overlap=True, ) params1.enable_chunked_context = True params2 = pickle.loads(pickle.dumps(params1)) assert params2 == params1 params1 = _tb.TrtGptModelOptionalParams() params2 = pickle.loads(pickle.dumps(params1)) assert params2 == params1 def test_Mpicomm(): size1 = _tb.MpiComm.getSize() rank1 = _tb.MpiComm.getRank() session_size = (size1 + 1) // 2 session_color = rank1 // session_size session_rank = rank1 % session_size _tb.MpiComm.split(session_color, session_rank) rank2 = _tb.MpiComm.getRank() size2 = _tb.MpiComm.getSize() assert rank2 == session_rank assert size2 == session_size def test_SamplingConfig_pickle(): config = _tb.SamplingConfig() config.beam_width = 2 config.temperature = [1.0, 2.0] config.top_k = [1, 2] config.top_p = [0.1, 0.2] config.random_seed = [1, 2] config.repetition_penalty = [1.0, 2.0] config.presence_penalty = [1.0, 2.0] config.frequency_penalty = [1.0, 2.0] config.length_penalty = [1.0, 2.0] config.early_stopping = [1, 2] config.top_p_decay = [1.0, 2.0] config.top_p_min = [1.0, 2.0] config.top_p_reset_ids = [1, 2] config.beam_search_diversity_rate = [1.0, 2.0] config1 = pickle.loads(pickle.dumps(config)) assert config1 == config