biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[BugFix] Change support no act and mul for marlin (#34088)

Browse Source 
Signed-off-by: Tomer Natan <tbarnatan@computelab-frontend-8.nvidia.com>
Co-authored-by: Tomer Natan <tbarnatan@computelab-frontend-8.nvidia.com>
This commit is contained in:
TomerBN-Nvidia
2026-02-08 19:18:22 +02:00
committed by GitHub
parent 179ae7da8f
commit a263aa6140
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/fused_moe/fused_marlin_moe.py
View File

@@ -561,7 +561,7 @@ class MarlinExpertsBase(mk.FusedMoEPermuteExpertsUnpermute):
@staticmethod @staticmethod
def _supports_no_act_and_mul() -> bool: def _supports_no_act_and_mul() -> bool:
return False return True
@staticmethod @staticmethod
def _supports_quant_scheme( def _supports_quant_scheme(