o ii @sJddlmZmZddlmZddlmZddlmZeGdddZ dS))AnyLiteral)field_validator)config)AttentionBackendEnumc@seZdZUdZdZedBed< dZeddBed< dZ e ed< dZ e ed < dZ e ed < d Ze ed < dZe dBed < dZe ed< dZe ed< defddZedddededefddZdS)AttentionConfigz/Configuration for attention mechanisms in vLLM.Nbackend)flash_attn_versionFuse_prefill_decode_attention (flash_attn_max_num_splits_for_cuda_graphuse_cudnn_prefillT"use_trtllm_ragged_deepseek_prefilluse_trtllm_attentiondisable_flashinfer_prefill!disable_flashinfer_q_quantizationreturncCs&ddlm}m}g}|||}||S)a$ Provide a hash that uniquely identifies all the configs that affect the structure of the computation graph from input ids/embeddings to the final hidden states, excluding anything before input ids/embeddings and after the final hidden states. r)get_hash_factors hash_factors)vllm.config.utilsrr)selfrrignored_factorsfactorsrK/home/ubuntu/vllm_env/lib/python3.10/site-packages/vllm/config/attention.py compute_hash.s zAttentionConfig.compute_hashbefore)modevaluecCst|tr t|S|S)z6Enable parsing of the `backend` enum type from string.) isinstancestrrupper)clsr rrrvalidate_backend_before<s  z'AttentionConfig.validate_backend_before)__name__ __module__ __qualname____doc__rr__annotations__r rr boolrintrrrrrr"rr classmethodrr%rrrrr s0        rN) typingrrpydanticrrr#vllm.v1.attention.backends.registryrrrrrrs