xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-05 05:24:37 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix vLLM x torch.compile config caching (#16491)

Browse Source 
Signed-off-by: rzou <zou3519@gmail.com>
This commit is contained in:
Richard Zou 2025-04-15 02:11:11 -04:00 committed by GitHub
parent b4fe16c75b
commit b590adfdc1
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
vllm/config.py
View File

@ -298,12 +298,18 @@ class ModelConfig:
factors.append(self.quantization) factors.append(self.quantization)
factors.append(self.revision) factors.append(self.revision)
factors.append(self.code_revision) factors.append(self.code_revision)
factors.append(self.max_model_len)
factors.append(self.max_logprobs)
factors.append(self.disable_sliding_window)
factors.append(self.trust_remote_code) factors.append(self.trust_remote_code)
factors.append(self.mm_processor_kwargs)
factors.append(self.generation_config)
factors.append(self.model_impl)
factors.append(self.override_generation_config)
factors.append(self.rope_scaling) factors.append(self.rope_scaling)
factors.append(self.rope_theta) factors.append(self.rope_theta)
# rope cos/sin cache depends on the max_position_embeddings # hf_config can control how the model looks!
factors.append( factors.append(self.hf_config.to_json_string())
getattr(self.hf_config, "max_position_embeddings", "None"))
return hashlib.sha256(str(factors).encode()).hexdigest() return hashlib.sha256(str(factors).encode()).hexdigest()
def __init__( def __init__(