xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-07-24 05:07:11 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Disable gptq_bitblas for <SM80 to fix GPTQ on V100/T4 (#17541)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Michael Goin 2025-05-01 18:59:35 -06:00 committed by GitHub
parent 39c0813a7f
commit 24aebae177
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/quantization/gptq_bitblas.py
View File

@ -134,7 +134,7 @@ class GPTQBitBLASConfig(QuantizationConfig):
@classmethod @classmethod
def get_min_capability(cls) -> int: def get_min_capability(cls) -> int:
return 70 return 80
@classmethod @classmethod
def get_config_filenames(cls) -> List[str]: def get_config_filenames(cls) -> List[str]: