EngineX-MetaX/enginex-c_series-vllm

Go to file

wangjing ce688181e6 [gpt-oss] Add gpt-oss mxfp4 support

2025-08-25 17:41:34 +08:00

[gpt-oss] Add gpt-oss mxfp4 support

2025-08-25 17:41:34 +08:00

.gitignore

[gpt-oss] Add gpt-oss mxfp4 support

2025-08-25 17:41:34 +08:00

code_generator.py

add Dockerfile

2025-08-18 02:43:10 +00:00

Dockerfile

add Dockerfile

2025-08-18 02:43:10 +00:00

README.md

[gpt-oss] Add gpt-oss mxfp4 support

2025-08-25 17:41:34 +08:00

README.md

metax-c500-vllm

支持 gpt-oss：将 vllm 目录覆盖到镜像中的 /opt/conda/lib/python3.10/site-packages/vllm。运行gpt-oss时需指定VLLM_ATTENTION_BACKEND=TRITON_ATTN_VLLM_V1
将 code_generator.py 覆盖到镜像中的 /opt/conda/lib/python3.10/site-packages/triton/compiler/code_generator.py