Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Misc]: vllm-ascend 推理速度非常慢 #171

Open
ryys1122 opened this issue Feb 26, 2025 · 2 comments
Open

[Misc]: vllm-ascend 推理速度非常慢 #171

ryys1122 opened this issue Feb 26, 2025 · 2 comments

Comments

@ryys1122
Copy link

Anything you want to discuss about vllm on ascend.

vllm-asend 部署成功后,使用4张910B,运行推理服务。

VLLM_USE_MODELSCOPE=true NPU_VISIBLE_DEVICES=4,5,6,7 ASCEND_RT_VISIBLE_DEVICES=4,5,6,7 vllm serve --tensor-parallel-size 4 deepseek-ai/DeepSeek-R1-Distill-Qwen-32B

测试推理服务

curl http://localhost:8000/v1/completions -H "Content-Type: application/json" -d '{
"model": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",
"prompt": "描述一下北京的秋天",
"max_tokens": 512
}'

将近3分钟才返回结果。

@caolicaoli
Copy link

我这里4卡运行llama70b,每秒9token,100并发下每秒700token。

@ryys1122
Copy link
Author

ryys1122 commented Mar 6, 2025

我这里4卡运行llama70b,每秒9token,100并发下每秒700token。

运行的命令和参数是什么样的呢?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

3 participants