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

[MoE] Move FlashInfer CuteDSL experts into fused_moe/experts/ (#37759)

Browse Source 
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
This commit is contained in:
Robert Shaw
2026-03-21 19:15:16 -04:00
committed by GitHub
parent eeee5b262d
commit 6b2fa3a762
3 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

0
vllm/model_executor/layers/fused_moe/flashinfer_cutedsl_moe.py → vllm/model_executor/layers/fused_moe/experts/flashinfer_cutedsl_moe.py
View File

2
vllm/model_executor/layers/fused_moe/oracle/nvfp4.py
View File

@@ -86,7 +86,7 @@ def backend_to_kernel_cls(
return [FlashInferExperts]
elif backend == NvFp4MoeBackend.FLASHINFER_CUTEDSL:
from vllm.model_executor.layers.fused_moe.flashinfer_cutedsl_moe import (
from vllm.model_executor.layers.fused_moe.experts.flashinfer_cutedsl_moe import ( # noqa: E501
FlashInferCuteDSLExperts,
)