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.

Sign up for GitHub

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

Jump to bottom

supporting PagedAttention in optimized models #735

Open
YYue000 opened this issue Nov 21, 2024 · 1 comment
Open

supporting PagedAttention in optimized models #735

YYue000 opened this issue Nov 21, 2024 · 1 comment
Assignees
@devpramod

Comments

@YYue000
Copy link

YYue000 commented Nov 21, 2024

Describe the issue

I notice there are implementations of PagedAttention for the CPU backend. However, these ops are never used in ipex.llm.optimize. Is there any opportunity to support the PagedAttention version of compiled models?
@devpramod devpramod self-assigned this Nov 26, 2024
@devpramod
Copy link
Contributor

Hi @YYue000
Which op are you referring to here? Could you share the implementation?
By "PagedAttention version of compiled models", do you mean with torch.compile?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@devpramod devpramod

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@YYue000 @devpramod