vllm/vllm/model_executor/layers/fused_moe at f243abc92d805b366ca00911d6ae5572faa39bff - vllm

configs

Add Fused MoE Triton kernels for GLM-4.5-Air, GLM-4.5v, GLM-4.6v on 2x RTX Pro 6000 (#31407 )

2025-12-28 08:38:33 -08:00

oracle

[MoE Refactor][16/N] Apply Refactor to NVFP4 (#31692 )

2026-01-08 03:46:27 +00:00

__init__.py

[MoE Refactoring][Bugfix]Wrap WNA16 Triton kernel into mk and change compressed tensor kernel selection (#31752 )

2026-01-08 19:01:30 -05:00

all2all_utils.py

[Bugfix] Fix Broken ModelOpt NVFP4 MoE (#31742 )

2026-01-05 23:18:38 +00:00

batched_deep_gemm_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

config.py

Revert "feat(moe): Add is_act_and_mul=False support for Triton MoE kernels" (#31978 )

2026-01-08 11:31:53 -08:00

cpu_fused_moe.py

[Misc] Fix Current vLLM config is not set. warnings, assert to avoid issues in the future (#31747 )

2026-01-08 15:20:49 -08:00

cutlass_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

deep_gemm_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

deep_gemm_utils.py

[Perf] Optimize deepgemm experts initialization, 3.9% TTFT improvement (#30494 )

2025-12-11 17:28:34 -05:00

deepep_ht_prepare_finalize.py

[Bug] Fix DBO IMA issue for DeepEPHT (#27666 )

2025-10-29 16:28:27 -04:00

deepep_ll_prepare_finalize.py

Add back missing DeepEP LL params (#31911 )

2026-01-07 17:47:54 -05:00

fallback.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

flashinfer_cutedsl_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

flashinfer_cutlass_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

flashinfer_cutlass_prepare_finalize.py

[MoE Refactor][16/N] Apply Refactor to NVFP4 (#31692 )

2026-01-08 03:46:27 +00:00

flashinfer_trtllm_moe.py

[MoE Refactor][15/N] Apply Refactor to Fp8 (#31415 )

2026-01-07 19:42:33 -05:00

fused_batched_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

fused_marlin_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

fused_moe_method_base.py

[Misc][Refactor] Add FusedMoERouter object (#30519 )

2026-01-08 20:52:55 +00:00

fused_moe_modular_method.py

[Doc] Add developer guide for CustomOp (#30886 )

2026-01-09 16:21:11 +00:00

fused_moe_router.py

[Misc][Refactor] Add FusedMoERouter object (#30519 )

2026-01-08 20:52:55 +00:00

fused_moe.py

[Kernel][MoE] fix computation order of MoE weight multiplication and improve flow (#31962 )

2026-01-12 17:17:30 -05:00

gpt_oss_triton_kernels_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

layer.py

[Feature] Support recording expert indices for rollout router replay (#28284 )

2026-01-12 06:23:04 -08:00

modular_kernel.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

moe_align_block_size.py

[Kernel][MoE] optimize moe_align_block_size (#29642 )

2025-12-07 01:58:47 -08:00

moe_permute_unpermute.py

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633 )

2025-10-12 09:51:31 -07:00

moe_torch_iterative.py

Convert formatting to use ruff instead of yapf + isort (#26247 )

2025-10-05 07:06:22 -07:00

pplx_prepare_finalize.py

[Chore]:Extract math and argparse utilities to separate modules (#27188 )

2025-10-26 04:03:32 -07:00

prepare_finalize.py

[MoE Refactor][7/N] AITER MK (#31102 )

2025-12-22 16:42:58 -07:00

rocm_aiter_fused_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

routed_experts_capturer.py

[Feature] Support recording expert indices for rollout router replay (#28284 )

2026-01-12 06:23:04 -08:00

routing_simulator.py

[Bugfix] Fix MoE Routing Simulation (#28002 )

2025-11-03 22:26:49 +00:00

shared_fused_moe.py

[Feat] Enable eplb with default all2all backend (#30559 )

2025-12-16 10:33:52 -05:00

topk_weight_and_reduce.py

Update Optional[x] -> x | None and Union[x, y] to x | y (#26633 )

2025-10-12 09:51:31 -07:00

triton_cutlass_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

triton_deep_gemm_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

trtllm_moe.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

unquantized_fused_moe_method.py

[Doc] Add developer guide for CustomOp (#30886 )

2026-01-09 16:21:11 +00:00

utils.py

[FIX] Add NO_MUL activation support for modular kernel path (#31528 )

2026-01-12 11:55:49 -05:00

zero_expert_fused_moe.py

[MoE Refactor][5/N] Isolate zero expert to LongCatFlash (#28891 )

2025-12-20 18:22:04 +00:00