Elvir Crnčević b9b401b2ef [Kernel] Fuse DeepEP-v2 decode globalize+align into one launch (#53911) ...

Signed-off-by: Elvir Crncevic <elvircrn@gmail.com>

2026-10-13 11:10:38 -04:00

..

__init__.py

…

aiter_mxfp4_w4a8_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

aiter_mxfp4_w4a16_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

aiter_mxfp8_moe.py

[Feature] Support batch invariance on ROCm (#52231)

2026-10-12 10:53:26 -05:00

batched_deep_gemm_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

cpu_int4_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

cpu_moe.py

[Refactor] Remove dead code multiple places (#60120)

2026-10-10 15:41:10 +00:00

cutlass_moe.py

[Bugfix][MoE] Apply expert_map remapping for CUTLASS FP4 MOE expert parallel (#59042)

2026-10-12 10:02:56 -04:00

deep_gemm_moe.py

[Perf][MoE] Skip padded work in block-FP8 DeepGEMM experts (#59128)

2026-10-06 18:49:57 -05:00

fallback.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

flashinfer_b12x_moe.py

[Bugfix][MoE] Support GELU tanh in FlashInfer B12x MoE (#51819)

2026-08-11 15:07:58 +00:00

flashinfer_cutedsl_batched_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

flashinfer_cutedsl_moe.py

[Quantization] Add per-token NVFP4 CuTe-DSL MoE backend (#50030)

2026-09-30 10:42:03 +00:00

flashinfer_cutlass_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

flashinfer_moe_ep.py

[feat] FlashInfer CuteDSL MegaMoE integration (#54049)

2026-10-02 15:43:33 -04:00

fused_batched_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

fused_humming_moe.py

[Kernel] Fuse DeepEP-v2 decode globalize+align into one launch (#53911)

2026-10-13 11:10:38 -04:00

gpt_oss_triton_kernels_moe.py

[Bugfix][MoE] Apply expert_map remapping for CUTLASS FP4 MOE expert parallel (#59042)

2026-10-12 10:02:56 -04:00

int4_emulation_moe.py

[feature]Add int4 quantization support for emulation moe backend (#48451)

2026-07-16 02:00:12 +00:00

lora_context.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

lora_experts_mixin.py

[Bugfix][LoRA] Fix GPT-OSS gate/up interleaving with mixed MoE LoRA formats (#60812)

2026-10-10 12:52:51 +08:00

marlin_moe.py

[Core][Kernel] Share persistent workspaces for Marlin and Humming (#57421)

2026-09-20 17:28:33 +08:00

moonep_experts.py

[Distributed][MoonEP] BF16 integration of MoonEP balanced EP backend (#52101)

2026-09-18 10:34:01 -04:00

mxfp8_emulation_moe.py

[MoE] Share apply_moe_activation support metadata (#44359)

2026-08-05 12:05:47 -07:00

mxfp8_native_moe.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

nvfp4_emulation_moe.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

ocp_mx_emulation_moe.py

[Feature] Support batch invariance on ROCm (#52231)

2026-10-12 10:53:26 -05:00

rdna3_moe.py

[Refactor][ROCm] Migrate the RDNA3 W4A16 MoE to the oracle/experts pa… (#55522)

2026-09-12 21:20:26 +08:00

rocm_aiter_moe.py

[ROCm][MoE] Split AiterExperts by weight-quantization key (#60322)

2026-10-13 00:13:33 -05:00

triton_cutlass_moe.py

[MoE Refactor] Move remaining experts classes to experts directory (#42334)

2026-05-12 09:19:46 -04:00

triton_deep_gemm_moe.py

[MoE Refactor] Move remaining experts classes to experts directory (#42334)

2026-05-12 09:19:46 -04:00

triton_moe.py

[Perf][MoE] Skip top-k slots routed to non-local experts in TritonExp… (#58051)

2026-09-23 21:03:21 +08:00

trtllm_bf16_moe.py

[Bugfix][MoE] Allow MiniMax2 routing in TRTLLM BF16 monolithic backend (#59571)

2026-10-05 16:24:14 -07:00

trtllm_fp8_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

trtllm_lora_moe.py

[MoE] Encapsulate TRT-LLM BF16 weight layout handling (#57405)

2026-09-17 17:47:30 -04:00

trtllm_mxfp4_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

trtllm_mxint4_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

trtllm_nvfp4_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00

xpu_moe.py

[XPU]Fix the accuracy issue when meet topk_ids=-1 on DP+EP scenarios (#57787)

2026-10-09 20:01:54 +08:00

zentorch_moe.py

[Bugfix] Bind routed-experts capture to the MoE layer, not the kernel (#59455)

2026-10-02 15:50:38 +00:00