Skip to content

fix the ops issue when integrate vllm hpu extension to TGI #116

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

Merged
merged 9 commits into from
Apr 29, 2025

Conversation

sywangyi
Copy link
Contributor

No description provided.

@sywangyi
Copy link
Contributor Author

Hi, I'd like to integrate ops level optimization like flat-pa/MOE to TGI. I found some issue during integration, so I give a fix. please help review and merge. Thanks ! MOE is founding during the enabling of https://huggingface.co/microsoft/Phi-3.5-MoE-instruct

@sywangyi
Copy link
Contributor Author

fix in pipelined_pa is found in TGI benchmark where only block 0 is used

@sywangyi
Copy link
Contributor Author

@yao-matrix

fmt and remove unused logger
@sywangyi
Copy link
Contributor Author

@mswiniarsk done for the review comment

Copy link
Contributor

@mswiniarsk mswiniarsk left a comment

Choose a reason for hiding this comment

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

LGTM

@michalkuligowski michalkuligowski merged commit 5b195ba into HabanaAI:main Apr 29, 2025
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.

3 participants