[MoE] Move DEEP_GEMM into experts/ subdirectory (#39005)

Signed-off-by: Jackmin801 <ongjackm@gmail.com>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
This commit is contained in:
Jackmin801
2026-04-08 19:23:08 +00:00
committed by GitHub
co-authored by Robert Shaw Robert Shaw
parent 8477fe427d
commit a776a48b1c
14 changed files with 24 additions and 20 deletions
@@ -77,16 +77,18 @@ __all__ = [
if HAS_TRITON:
# import to register the custom ops
from vllm.model_executor.layers.fused_moe.batched_deep_gemm_moe import (
BatchedDeepGemmExperts,
)
from vllm.model_executor.layers.fused_moe.cutlass_moe import (
CutlassBatchedExpertsFp8,
CutlassExpertsFp8,
CutlassExpertsW4A8Fp8,
cutlass_moe_w4a8_fp8,
)
from vllm.model_executor.layers.fused_moe.deep_gemm_moe import DeepGemmExperts
from vllm.model_executor.layers.fused_moe.experts.batched_deep_gemm_moe import (
BatchedDeepGemmExperts,
)
from vllm.model_executor.layers.fused_moe.experts.deep_gemm_moe import (
DeepGemmExperts,
)
from vllm.model_executor.layers.fused_moe.fused_batched_moe import (
BatchedTritonExperts,
)
@@ -131,7 +131,7 @@ def backend_to_kernel_cls(
return [TritonOrDeepGemmExperts]
elif backend == Fp8MoeBackend.BATCHED_DEEPGEMM:
from vllm.model_executor.layers.fused_moe.batched_deep_gemm_moe import (
from vllm.model_executor.layers.fused_moe.experts.batched_deep_gemm_moe import (
BatchedDeepGemmExperts,
)
@@ -9,7 +9,7 @@ from vllm.model_executor.layers.fused_moe.config import (
FusedMoEConfig,
FusedMoEQuantConfig,
)
from vllm.model_executor.layers.fused_moe.deep_gemm_moe import (
from vllm.model_executor.layers.fused_moe.experts.deep_gemm_moe import (
DeepGemmExperts,
_valid_deep_gemm,
_valid_deep_gemm_shape,
@@ -11,8 +11,8 @@ from tqdm import tqdm
import vllm.envs as envs
from vllm.distributed.parallel_state import get_dp_group, is_global_first_rank
from vllm.model_executor.layers.fused_moe.deep_gemm_moe import DeepGemmExperts
from vllm.model_executor.layers.fused_moe.deep_gemm_utils import compute_aligned_M
from vllm.model_executor.layers.fused_moe.experts.deep_gemm_moe import DeepGemmExperts
from vllm.model_executor.layers.fused_moe.layer import FusedMoE
from vllm.model_executor.layers.fused_moe.triton_deep_gemm_moe import (
TritonOrDeepGemmExperts,