EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[Doc] Update the Pangu Pro MoE tutorials. (#3651)

Browse Source 
### What this PR does / why we need it?
Update the Pangu Pro MoE tutorials.
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.11.0rc3
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0

Signed-off-by: menogrey <1299267905@qq.com>
This commit is contained in:
zhangyiming
2025-10-23 20:41:47 +08:00
committed by GitHub
parent aeddf4261a
commit ebfd09a075
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/source/tutorials/multi_npu_moge.md
View File

@@ -51,6 +51,7 @@ vllm serve /path/to/pangu-pro-moe-model \
--tensor-parallel-size 4 \ --tensor-parallel-size 4 \
--enable-expert-parallel \ --enable-expert-parallel \
--trust-remote-code \ --trust-remote-code \
--max_model_len=1024 \
--enforce-eager --enforce-eager
``` ```
@@ -217,6 +218,7 @@ if __name__ == "__main__":
llm = LLM(model="/path/to/pangu-pro-moe-model", llm = LLM(model="/path/to/pangu-pro-moe-model",
tensor_parallel_size=4, tensor_parallel_size=4,
enable_expert_parallel=True,
distributed_executor_backend="mp", distributed_executor_backend="mp",
max_model_len=1024, max_model_len=1024,
trust_remote_code=True, trust_remote_code=True,