Wei 0efdb5c3ba

[gpt-oss] Cache permute indices for faster MXFP4 MoE layer loading (#24154) ...

Signed-off-by: Wei Wei <wwei6@meta.com>

2025-09-10 04:27:53 +00:00

..

modular_kernel_tools

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

__init__.py

[Kernel] DeepEP dispatch-combine kernel integration (#18434)

2025-06-03 12:30:02 -07:00

parallel_utils.py

[Misc] Add unit tests for MoE ModularKernel combinations + Profiling utility (#20449)

2025-07-11 07:51:46 -07:00

test_batched_deepgemm.py

[Test] Add Unit Test for Batched DeepGEMM (#21559)

2025-08-02 10:45:46 +08:00

test_batched_moe.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_block_fp8.py

[Feature] Add Hopper DeepGEMM E8M0 for DeepSeekV3.1 scale_fmt (#23666)

2025-08-27 14:09:08 +00:00

test_block_int8.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_count_expert_num_tokens.py

[CI Perf] Prune tests in tests/kernels/moe/ (#22939)

2025-08-14 21:33:42 -06:00

test_cutlass_grouped_gemm.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_cutlass_moe.py

[Kernel] CUTLASS MoE FP8: Integrate cuda moe permute/unpermute (#23045)

2025-08-20 10:35:26 -04:00

test_deepep_deepgemm_moe.py

[Doc]: fix typos in Python comments (#24077)

2025-09-02 02:38:55 -07:00

test_deepep_moe.py

[CI Fix] Pin deepep and pplx tags in tools/ep_kernels/, gate multigpu tests (#23568)

2025-08-25 18:29:00 -07:00

test_deepgemm.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_flashinfer_moe.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_flashinfer.py

[NVIDIA] Add SM100 Flashinfer Cutlass MoE fp8 backend (#22357)

2025-08-19 18:01:53 -04:00

test_gpt_oss_triton_kernels.py

[Bugfix] Fix CI moe kernel failure (#22556)

2025-08-09 00:03:29 -07:00

test_grouped_topk.py

[Kernel] Add fused grouped_topk kernel for MoE (#23274)

2025-08-25 11:47:52 -07:00

test_modular_kernel_combinations.py

[CI Fix] Pin deepep and pplx tags in tools/ep_kernels/, gate multigpu tests (#23568)

2025-08-25 18:29:00 -07:00

test_moe_align_block_size.py

[CI Perf] Prune tests in tests/kernels/moe/ (#22939)

2025-08-14 21:33:42 -06:00

test_moe_permute_unpermute.py

[Kernel] CUTLASS MoE FP8: Integrate cuda moe permute/unpermute (#23045)

2025-08-20 10:35:26 -04:00

test_moe.py

[Bugfix] Fix test_mixtral_moe (#24371)

2025-09-06 09:32:03 -07:00

test_mxfp4_moe.py

[gpt-oss] Cache permute indices for faster MXFP4 MoE layer loading (#24154)

2025-09-10 04:27:53 +00:00

test_nvfp4_moe.py

[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)

2025-08-15 14:46:00 -04:00

test_pplx_cutlass_moe.py

[CI Fix] Pin deepep and pplx tags in tools/ep_kernels/, gate multigpu tests (#23568)

2025-08-25 18:29:00 -07:00

test_pplx_moe.py

[Kernels] Overlap shared experts with send/recv (#23273)

2025-09-03 12:35:18 -04:00

test_rocm_aiter_topk.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

test_silu_mul_fp8_quant_deep_gemm.py

[Perf] Small optimizations for silu_mul_fp8_quant_deep_gemm (#23265)

2025-08-21 17:56:15 -04:00

test_triton_moe_ptpc_fp8.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

utils.py

[Kernels] Overlap shared experts with send/recv (#23273)

2025-09-03 12:35:18 -04:00