Fr4nk1in 4bd2952a37 feat: add dp attention support for Qwen 2/3 MoE models, fixes #6088 (#6121) ...

Co-authored-by: King.Zevin <zevin@mail.ustc.edu.cn>
Co-authored-by: Yi Zhang <1109276519@qq.com>

2025-05-16 14:44:10 -07:00

..

attention

[Feat] Support FlashMLA backend with MTP and FP8 KV cache (#6109)

2025-05-15 00:48:09 -07:00

moe

[2/2] Add python wrapper for CUTLASS FP8 Blockscale MoE Kernel. (#5694)

2025-05-16 13:14:07 -07:00

quantization

[2/2] Add python wrapper for CUTLASS FP8 Blockscale MoE Kernel. (#5694)

2025-05-16 13:14:07 -07:00

activation.py

[misc] remove is_cuda_available (#5319)

2025-04-20 18:16:51 -07:00

dp_attention.py

feat: add dp attention support for Qwen 2/3 MoE models, fixes #6088 (#6121)

2025-05-16 14:44:10 -07:00

elementwise.py

Fix run time error in ROCm platform (#5147)

2025-04-07 22:49:40 -07:00

layernorm.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

linear.py

[Model Support] unsloth/Phi-4-mini bnb model (#4982)

2025-04-16 19:58:20 -07:00

logits_processor.py

Fix two issues related to --moe-dense-tp-size=1 (#5657)

2025-05-12 23:51:39 -07:00

parameter.py

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

pooler.py

[feature] support for roberta embedding models (#5730)

2025-04-26 18:47:06 -07:00

radix_attention.py

Fuse MLA set kv cache kernel (#5748)

2025-04-26 18:42:22 -07:00

rotary_embedding.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

sampler.py

Improve structured outputs: fix race condition, server crash, metrics and style (#6188)

2025-05-11 08:36:16 -07:00

torchao_utils.py

Support loading of larger models with on-the-fly quantization (#3061)

2025-01-22 21:33:17 -08:00

utils.py

[PP] Add pipeline parallelism (#5724)

2025-04-30 18:18:07 -07:00

vocab_parallel_embedding.py

Performing Vocabulary Parallelism for LM Head across Attention TP Groups (#5558)

2025-05-11 23:36:29 -07:00