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

[Feature] Add env var VLLM_MOE_USE_DEEP_GEMM (#28422)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2025-11-10 21:29:48 -05:00
committed by GitHub
parent 39029d5192
commit de540c0354
4 changed files with 19 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/model_executor/layers/quantization/fp8.py
View File

@@ -158,7 +158,7 @@ def get_fp8_moe_backend(block_quant: bool) -> Fp8MoeBackend:
return Fp8MoeBackend.MARLIN
# deepGEMM on supported platforms with block-quantized weights
if envs.VLLM_USE_DEEP_GEMM and block_quant:
if envs.VLLM_USE_DEEP_GEMM and envs.VLLM_MOE_USE_DEEP_GEMM and block_quant:
if not has_deep_gemm():
logger.warning_once("DeepGEMM backend requested but not available.")
elif is_deep_gemm_supported():