RichardWooSJTU 9f0778f991 [Feature] Support EP prefill with num_worst_tokens (#6574) ...

* support num worst tokens

* support num worst tokens

* fix build error

* support num worst tokens: fix errors

* support num worst tokens: fix feild

* support num worst tokens: delete requiements

* replace permute and depermute op by pure cuda

* replace permute and depermute op by pure cuda

* fix ci

* fix op

* fix nan

* fix code style

---------

Co-authored-by: YuBaoku <49938469+EmmonsCurse@users.noreply.github.com>

2026-03-11 17:09:07 +08:00

..

ops

[Feature] Support Ernie FP8 on sm100 ( the fixed version) (#6304)

2026-02-03 17:47:38 +08:00

__init__.py

[Feature] Support NVFP4 MoE on SM100 (#6003)

2026-01-29 14:16:07 +08:00

block_wise_fp8.py

[Feature] Support EP prefill with num_worst_tokens (#6574)

2026-03-11 17:09:07 +08:00

fp8_utils.py

[Models][OP][Optimization] Support DeepSeek-v3.2 model, integrate DSA & Indexer architecture with FlashMLA/DeepGEMM (#6689)

2026-03-10 15:05:14 +08:00

kv_cache.py

[Intel HPU] enable tensor_wise_fp8 (#5324)

2025-12-17 16:45:03 +08:00

mix_quant.py

support w4afp8 moe offline permute & load (#5613)

2025-12-22 15:12:57 +08:00

mxfp4.py

Support MXFP4 for GPT-OSS (#5435)

2026-01-22 14:21:01 +08:00

nvfp4.py

nvfp4 clean code (#6671)

2026-03-09 18:00:34 +08:00

quant_base.py

polish code with new pre-commit rule (#2923)

2025-07-19 23:19:27 +08:00

tensor_wise_fp8.py

[Intel HPU] enable tensor_wise_fp8 (#5324)

2025-12-17 16:45:03 +08:00

w4a8.py

[XPU] refactor moe ffn (#5501)

2025-12-18 14:14:05 +08:00

w4afp8.py

[Feature] support w4afp8 v1_loader and v0_loader(tp>1) (#5757)

2025-12-30 14:11:52 +08:00

w8a8.py

fix w8a8.py (#3733)

2025-09-03 10:57:26 +08:00

weight_only.py

[Models][OP][Optimization] Support DeepSeek-v3.2 model, integrate DSA & Indexer architecture with FlashMLA/DeepGEMM (#6689)

2026-03-10 15:05:14 +08:00

wfp8afp8.py

[Loader] support dummy load weight (#6169)

2026-01-26 13:58:53 +08:00

wint2.py

fix wint2 config (#4721)

2025-10-31 15:44:14 +08:00