[MoE] Move cutlass moe to fused_moe/experts/ (#40574)

Signed-off-by: Jackmin801 <ongjackm@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
This commit is contained in:
Jackmin801
2026-04-24 06:05:49 +00:00
committed by GitHub
co-authored by Claude
parent 9744b699ba
commit 079a4cf399
14 changed files with 20 additions and 18 deletions
@@ -367,7 +367,9 @@ else:
CutlassExpertsFp8 = None
if cutlass_fp4_supported():
from vllm.model_executor.layers.fused_moe.cutlass_moe import CutlassExpertsFp4
from vllm.model_executor.layers.fused_moe.experts.cutlass_moe import (
CutlassExpertsFp4,
)
register_experts(
CutlassExpertsFp4,
+1 -1
View File
@@ -21,7 +21,7 @@ from vllm.model_executor.layers.fused_moe.config import (
FusedMoEQuantConfig,
fp8_w8a8_moe_quant_config,
)
from vllm.model_executor.layers.fused_moe.cutlass_moe import (
from vllm.model_executor.layers.fused_moe.experts.cutlass_moe import (
CutlassExpertsFp8,
run_cutlass_moe_fp8,
)
+1 -1
View File
@@ -19,7 +19,7 @@ from vllm.model_executor.layers.fused_moe.all2all_utils import (
maybe_make_prepare_finalize,
)
from vllm.model_executor.layers.fused_moe.config import nvfp4_moe_quant_config
from vllm.model_executor.layers.fused_moe.cutlass_moe import (
from vllm.model_executor.layers.fused_moe.experts.cutlass_moe import (
CutlassExpertsFp4,
)
from vllm.model_executor.layers.fused_moe.prepare_finalize import (