周周周 17b414c2df MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

..

attention

enable dcu ci (#3402)

2025-08-29 10:23:08 +08:00

backends

enable dcu ci (#3402)

2025-08-29 10:23:08 +08:00

moe

add input_processor plugin (#3657)

2025-08-28 22:53:57 +08:00

quantization

MoE Default use triton's blockwise fp8 in TP Case (#3678)

2025-08-29 11:07:30 +08:00

sample

[Feature] Add temp_scaled_logprobs and top_p_normalized_logprobs parameters for logits and logprobs post processing (#3552)

2025-08-25 14:11:49 +08:00

__init__.py

[LLM] First commit the llm deployment code

2025-06-09 19:20:15 +08:00

activation.py

[Polish Code] Remove useless notes

2025-08-14 14:04:52 +08:00

embeddings.py

Supports DP+TP+EP hybrid parallel deployment strategy (#3489)

2025-08-26 00:04:01 -07:00

linear.py

fix qwen3 235B tp 8 (#3697)

2025-08-28 23:46:25 +08:00

lm_head.py

[Precision] Support lm_head layer running in float32 (#3597)

2025-08-27 11:34:53 +08:00

mtp_linear.py

support tmp (#3675)

2025-08-28 19:42:32 +08:00

normalization.py

adaptive rms_norm's dtype (#3617)

2025-08-26 15:29:15 +08:00

rotary_embedding.py

[MetaxGPU] Support FastDeploy on metax gpu (#3241)

2025-08-13 11:11:54 +08:00

utils.py

[V1 Loader] support weight_only (#3413)

2025-08-23 13:13:41 +08:00