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

Add compute capability 8.9 to default targets #829

Merged
merged 2 commits into from
Aug 22, 2023
Merged

Add compute capability 8.9 to default targets #829

merged 2 commits into from
Aug 22, 2023

Conversation

WoosukKwon
Copy link
Collaborator

@WoosukKwon WoosukKwon commented Aug 22, 2023

This PR adds compute capability 8.9 to the default target compute capabilities. While SM89 GPUs (such as L4 and RTX4090) can run the code generated for SM80 or SM89, the code targeting SM89 may achieve higher performance.

This PR also fixes #803

@WoosukKwon WoosukKwon changed the title Add compute capability 8.9 to targets Add compute capability 8.9 to default targets Aug 22, 2023
@WoosukKwon WoosukKwon requested a review from zhuohan123 August 22, 2023 14:31
Copy link
Member

@zhuohan123 zhuohan123 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM!

@WoosukKwon WoosukKwon merged commit a41c204 into main Aug 22, 2023
@WoosukKwon WoosukKwon deleted the add89 branch August 22, 2023 22:28
randxie pushed a commit to randxie/vllm that referenced this pull request Aug 29, 2023
liuyanyi pushed a commit to liuyanyi/vllm that referenced this pull request Sep 12, 2023
hongxiayang pushed a commit to hongxiayang/vllm that referenced this pull request Feb 13, 2024
sjchoi1 pushed a commit to casys-kaist-internal/vllm that referenced this pull request May 7, 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.

Issue Building vllm Wheel: Unsupported GPU Architecture 'compute_89'
2 participants