Index A | B | C | D | E | F | G | H | I | K | L | M | N | O | P | R | S | T | U | V | W A abort() (vllm.AsyncLLMEngine method) abort_request() (vllm.LLMEngine method) add_lora() (vllm.AsyncLLMEngine method) add_request() (vllm.LLMEngine method) additional_data (vllm.PoolingParams attribute) apply() (vllm.multimodal.processing.BaseMultiModalProcessor method) (vllm.multimodal.processing.EncDecMultiModalProcessor method) apply_model() (vllm.LLM method) as_classification_model() (in module vllm.model_executor.models.adapters) as_embedding_model() (in module vllm.model_executor.models.adapters) as_reward_model() (in module vllm.model_executor.models.adapters) AsyncLLMEngine (class in vllm) AudioEmbeddingItems (class in vllm.multimodal.parse) AudioProcessorItems (class in vllm.multimodal.parse) B BaseDummyInputsBuilder (class in vllm.multimodal.profiling) BaseMultiModalProcessor (class in vllm.multimodal.processing) BaseProcessingInfo (class in vllm.multimodal.processing) batch() (vllm.multimodal.inputs.MultiModalKwargs static method) batched() (vllm.multimodal.inputs.MultiModalFieldConfig static method) beam_search() (vllm.LLM method) BoundPromptReplacement (class in vllm.multimodal.processing) C chat() (vllm.LLM method) check_health() (vllm.AsyncLLMEngine method) classify() (vllm.LLM method) clone() (vllm.PoolingParams method) (vllm.SamplingParams method) collective_rpc() (vllm.LLM method) compute_logits() (vllm.model_executor.models.interfaces_base.VllmModelForTextGeneration method) create_encoder_prompt() (vllm.multimodal.processing.EncDecMultiModalProcessor method) create_input_mapper() (vllm.multimodal.registry.MultiModalRegistry method) create_processor() (vllm.multimodal.registry.MultiModalRegistry method) D data (vllm.multimodal.inputs.MultiModalFieldElem attribute) DEPRECATE_INIT_POSARGS (vllm.LLM attribute) DEPRECATE_LEGACY (vllm.LLM attribute) DictEmbeddingItems (class in vllm.multimodal.parse) do_log_stats() (vllm.LLMEngine method) DO_VALIDATE_OUTPUT (vllm.LLMEngine attribute) DummyInputsBuilderFactory (class in vllm.multimodal.registry) E embed() (vllm.LLM method) EmbeddingItems (class in vllm.multimodal.parse) EncDecMultiModalProcessor (class in vllm.multimodal.processing) encode() (vllm.AsyncLLMEngine method) (vllm.LLM method) engine_step() (vllm.AsyncLLMEngine method) F features (vllm.multimodal.processing.PromptReplacementDetails attribute) field (vllm.multimodal.inputs.MultiModalFieldElem attribute) find_text_matches() (in module vllm.multimodal.processing) find_token_matches() (in module vllm.multimodal.processing) flat() (vllm.multimodal.inputs.MultiModalFieldConfig static method) flat_from_sizes() (vllm.multimodal.inputs.MultiModalFieldConfig static method) forward() (vllm.model_executor.models.interfaces.SupportsPP method) from_engine_args() (vllm.AsyncLLMEngine class method) (vllm.LLMEngine class method) from_items() (vllm.multimodal.inputs.MultiModalKwargs static method) full (vllm.multimodal.processing.PromptReplacementDetails attribute) full_groupby_modality() (in module vllm.multimodal.processing) G generate() (vllm.AsyncLLMEngine method) (vllm.LLM method) get() (vllm.multimodal.parse.DictEmbeddingItems method) (vllm.multimodal.parse.EmbeddingItems method) (vllm.multimodal.parse.ModalityDataItems method) (vllm.multimodal.parse.ProcessorBatchItems method) get_all() (vllm.multimodal.parse.ModalityDataItems method) get_all_counts() (vllm.multimodal.parse.MultiModalDataItems method) get_count() (vllm.multimodal.parse.DictEmbeddingItems method) (vllm.multimodal.parse.EmbeddingItems method) (vllm.multimodal.parse.ModalityDataItems method) (vllm.multimodal.parse.MultiModalDataItems method) (vllm.multimodal.parse.ProcessorBatchItems method) get_decoding_config() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) get_dummy_processor_inputs() (vllm.multimodal.profiling.BaseDummyInputsBuilder method) get_hf_processor() (vllm.multimodal.processing.BaseProcessingInfo method) get_input_embeddings() (vllm.model_executor.models.interfaces.SupportsMultiModal method) get_input_preprocessor() (vllm.AsyncLLMEngine method) get_item() (vllm.multimodal.inputs.MultiModalKwargs method) get_item_count() (vllm.multimodal.inputs.MultiModalKwargs method) get_items() (vllm.multimodal.inputs.MultiModalKwargs method) (vllm.multimodal.parse.MultiModalDataItems method) get_lora_config() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) get_max_multimodal_tokens() (vllm.multimodal.registry.MultiModalRegistry method) get_max_tokens_by_modality() (vllm.multimodal.registry.MultiModalRegistry method) get_max_tokens_per_item_by_modality() (vllm.multimodal.registry.MultiModalRegistry method) get_max_tokens_per_item_by_nonzero_modality() (vllm.multimodal.registry.MultiModalRegistry method) get_mm_limits_per_prompt() (vllm.multimodal.registry.MultiModalRegistry method) get_mm_max_tokens_per_item() (vllm.multimodal.processing.BaseProcessingInfo method) get_model_config() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) get_multimodal_embeddings() (vllm.model_executor.models.interfaces.SupportsMultiModal method) get_num_unfinished_requests() (vllm.LLMEngine method) get_parallel_config() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) get_passthrough_data() (vllm.multimodal.parse.DictEmbeddingItems method) (vllm.multimodal.parse.EmbeddingItems method) (vllm.multimodal.parse.ModalityDataItems method) (vllm.multimodal.parse.ProcessorBatchItems method) get_processor_data() (vllm.multimodal.parse.DictEmbeddingItems method) (vllm.multimodal.parse.EmbeddingItems method) (vllm.multimodal.parse.ModalityDataItems method) (vllm.multimodal.parse.ProcessorBatchItems method) get_replacement() (vllm.multimodal.processing.BoundPromptReplacement method) get_scheduler_config() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) get_supported_mm_limits() (vllm.multimodal.processing.BaseProcessingInfo method) get_tokenizer() (vllm.AsyncLLMEngine method) H has_inner_state (vllm.model_executor.models.interfaces.HasInnerState attribute) has_processor() (vllm.multimodal.registry.MultiModalRegistry method) has_unfinished_requests() (vllm.LLMEngine method) has_unfinished_requests_for_virtual_engine() (vllm.LLMEngine method) HasInnerState (class in vllm.model_executor.models.interfaces) height (vllm.multimodal.parse.ImageSize attribute) I ImageEmbeddingItems (class in vllm.multimodal.parse) ImageProcessorItems (class in vllm.multimodal.parse) ImageSize (class in vllm.multimodal.parse) init_mm_limits_per_prompt() (vllm.multimodal.registry.MultiModalRegistry method) is_attention_free (vllm.model_executor.models.interfaces.IsAttentionFree attribute) is_hybrid (vllm.model_executor.models.interfaces.IsHybrid attribute) IsAttentionFree (class in vllm.model_executor.models.interfaces) IsHybrid (class in vllm.model_executor.models.interfaces) iter_token_matches() (in module vllm.multimodal.processing) K key (vllm.multimodal.inputs.MultiModalFieldElem attribute) L length (vllm.multimodal.inputs.PlaceholderRange attribute) LLM (class in vllm) LLMEngine (class in vllm) M make_empty_intermediate_tensors() (vllm.model_executor.models.interfaces.SupportsPP method) map_input() (vllm.multimodal.registry.MultiModalRegistry method) mm_hashes (vllm.multimodal.inputs.MultiModalInputs attribute) mm_kwargs (vllm.multimodal.inputs.MultiModalInputs attribute) mm_placeholders (vllm.multimodal.inputs.MultiModalInputs attribute) mm_processor_kwargs (vllm.inputs.TextPrompt attribute) (vllm.inputs.TokensPrompt attribute) modality (vllm.multimodal.inputs.MultiModalFieldElem attribute) (vllm.multimodal.processing.PromptReplacement attribute) ModalityDataItems (class in vllm.multimodal.parse) module vllm.engine vllm.model_executor.models.adapters vllm.model_executor.models.interfaces vllm.model_executor.models.interfaces_base vllm.multimodal.parse vllm.multimodal.processing vllm.multimodal.profiling vllm.multimodal.registry multi_modal_data (vllm.inputs.TextPrompt attribute) (vllm.inputs.TokensPrompt attribute) MULTIMODAL_REGISTRY (in module vllm.multimodal) MultiModalDataDict (in module vllm.multimodal.inputs) MultiModalDataItems (class in vllm.multimodal.parse) MultiModalDataParser (class in vllm.multimodal.parse) MultiModalFieldConfig (class in vllm.multimodal.inputs) MultiModalFieldElem (class in vllm.multimodal.inputs) MultiModalInputs (class in vllm.multimodal.inputs) MultiModalKwargs (class in vllm.multimodal.inputs) MultiModalKwargsItem (class in vllm.multimodal.inputs) MultiModalProcessorFactory (class in vllm.multimodal.registry) MultiModalProfiler (class in vllm.multimodal.profiling) MultiModalRegistry (class in vllm.multimodal.registry) N NestedTensors (in module vllm.multimodal.inputs) O offset (vllm.multimodal.inputs.PlaceholderRange attribute) P PlaceholderFeaturesInfo (class in vllm.multimodal.processing) PlaceholderRange (class in vllm.multimodal.inputs) pooler() (vllm.model_executor.models.interfaces_base.VllmModelForPooling method) PoolingParams (class in vllm) ProcessingInfoFactory (class in vllm.multimodal.registry) ProcessorBatchItems (class in vllm.multimodal.parse) ProcessorInputs (class in vllm.multimodal.profiling) prompt (vllm.inputs.TextPrompt attribute) (vllm.multimodal.inputs.MultiModalInputs attribute) prompt_token_ids (vllm.inputs.TokensPrompt attribute) (vllm.multimodal.inputs.MultiModalInputs attribute) PromptRepl (in module vllm.multimodal.processing) PromptReplacement (class in vllm.multimodal.processing) PromptReplacementDetails (class in vllm.multimodal.processing) PromptSeq (in module vllm.multimodal.processing) PromptType (in module vllm.inputs) R register_image_input_mapper() (vllm.multimodal.registry.MultiModalRegistry method) register_input_mapper() (vllm.multimodal.registry.MultiModalRegistry method) register_max_image_tokens() (vllm.multimodal.registry.MultiModalRegistry method) register_max_multimodal_tokens() (vllm.multimodal.registry.MultiModalRegistry method) register_plugin() (vllm.multimodal.registry.MultiModalRegistry method) register_processor() (vllm.multimodal.registry.MultiModalRegistry method) replace_text_matches() (in module vllm.multimodal.processing) replace_token_matches() (in module vllm.multimodal.processing) replacement (vllm.multimodal.processing.PromptReplacement attribute) reset_prefix_cache() (vllm.AsyncLLMEngine method) (vllm.LLMEngine method) run_engine_loop() (vllm.AsyncLLMEngine static method) S sample() (vllm.model_executor.models.interfaces_base.VllmModelForTextGeneration method) SamplingParams (class in vllm) score() (vllm.LLM method) shared() (vllm.multimodal.inputs.MultiModalFieldConfig static method) shutdown_background_loop() (vllm.AsyncLLMEngine method) sleep() (vllm.AsyncLLMEngine method) (vllm.LLM method) start_background_loop() (vllm.AsyncLLMEngine method) start_profile() (vllm.AsyncLLMEngine method) step() (vllm.LLMEngine method) stop_profile() (vllm.AsyncLLMEngine method) supports_lora (vllm.model_executor.models.interfaces.SupportsLoRA attribute) supports_multimodal (vllm.model_executor.models.interfaces.SupportsMultiModal attribute) supports_pp (vllm.model_executor.models.interfaces.SupportsPP attribute) SupportsCrossEncoding (class in vllm.model_executor.models.interfaces) SupportsLoRA (class in vllm.model_executor.models.interfaces) SupportsMultiModal (class in vllm.model_executor.models.interfaces) SupportsPP (class in vllm.model_executor.models.interfaces) SupportsQuant (class in vllm.model_executor.models.interfaces) SupportsTranscription (class in vllm.model_executor.models.interfaces) T target (vllm.multimodal.processing.BoundPromptReplacement property) (vllm.multimodal.processing.PromptReplacement attribute) TextPrompt (class in vllm.inputs) token_type_ids (vllm.inputs.TokensPrompt attribute) (vllm.multimodal.inputs.MultiModalInputs attribute) TokensPrompt (class in vllm.inputs) type (vllm.multimodal.inputs.MultiModalInputs attribute) U update_from_generation_config() (vllm.SamplingParams method) V VideoEmbeddingItems (class in vllm.multimodal.parse) VideoProcessorItems (class in vllm.multimodal.parse) vllm.engine module vllm.model_executor.models.adapters module vllm.model_executor.models.interfaces module vllm.model_executor.models.interfaces_base module vllm.multimodal.parse module vllm.multimodal.processing module vllm.multimodal.profiling module vllm.multimodal.registry module VllmModel (class in vllm.model_executor.models.interfaces_base) VllmModelForPooling (class in vllm.model_executor.models.interfaces_base) VllmModelForTextGeneration (class in vllm.model_executor.models.interfaces_base) W wake_up() (vllm.AsyncLLMEngine method) (vllm.LLM method) width (vllm.multimodal.parse.ImageSize attribute)
