bukejiyu 1539fd6056 [BugFix]Set default OMP_NUM_THREADS=3 and fix extra GPU memory usage in DeepSeek (#5219) ...

* fix bug

* update

* update

* update

* fix copy

* update

2025-11-28 14:22:04 +08:00

..

attention

[Feature] support flash_mask_attention backend (#5134)

2025-11-28 10:12:16 +08:00

backends

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

moe

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

pool

…

quantization

support moe offline quant (#5142)

2025-11-24 18:59:18 +08:00

sample

[Feature] ThreadPoolExecutor async fill_token_bitmask (#5083)

2025-11-19 10:04:16 +08:00

__init__.py

…

activation.py

…

embeddings.py

…

linear.py

[BugFix]Set default OMP_NUM_THREADS=3 and fix extra GPU memory usage in DeepSeek (#5219)

2025-11-28 14:22:04 +08:00

lm_head.py

[Optimization] Refine row parallel bias and nranks and moe all_reduce (#5247)

2025-11-26 05:09:09 -08:00

mtp_linear.py

[BugFix]fix v1 loader lm head fp32 (#5270)

2025-11-27 20:12:56 +08:00

normalization.py

…

pooler.py

…

rotary_embedding.py

…

utils.py

[BugFix]Set default OMP_NUM_THREADS=3 and fix extra GPU memory usage in DeepSeek (#5219)

2025-11-28 14:22:04 +08:00