freeliuzc 52eda7fdb3 [Feature][MTP]support new speculative decoding method named hybrid mtp with ngram (#3610)

2025-08-26 14:29:22 +08:00

..

ernie4_5_vl

[Executor] CUDAGraph support RL training (#3265)

2025-08-25 20:59:30 +08:00

__init__.py

…

deepseek_v3.py

qkv_a_proj horizontal fusion (#3591)

2025-08-26 14:25:57 +08:00

ernie4_5_moe.py

[Executor] CUDAGraph support RL training (#3265)

2025-08-25 20:59:30 +08:00

ernie4_5_mtp.py

[Feature][MTP]support new speculative decoding method named hybrid mtp with ngram (#3610)

2025-08-26 14:29:22 +08:00

model_base.py

…

qwen2.py

[Features] support hugging face qwen3 dense and qwen2 model (#3574)

2025-08-26 10:54:53 +08:00

qwen3.py

[Executor] CUDAGraph support RL training (#3265)

2025-08-25 20:59:30 +08:00

qwen3moe.py

[Executor] CUDAGraph support RL training (#3265)

2025-08-25 20:59:30 +08:00

tp_utils.py

…

utils.py

[V1 Loader] support weight_only (#3413)

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