Skip to content

Conversation

zhaochenyang20
Copy link

I added SGLang as an additional serving engine for Qwen models. Thus I changed the installation process to install vllm and sglang[all] together.

The comparison on my local machine is attached here:

image

image

All in all, SGLang is as well as vllm considering the accuracy and much faster as we expected. Thanks!

@zhaochenyang20 zhaochenyang20 changed the title Support As Serving Engine Support SGLang As Serving Engine Nov 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant