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

Enable the flaggem on triton-xpu backend and benchmark the performance on several oob model #2976

Open

2 of 4 tasks

zhczhong opened this issue Dec 10, 2024 · 0 comments

Assignees

Labels

performance research

Milestone

4.3 [Performance]...

Member

zhczhong commented Dec 10, 2024 •

edited

Loading

https://github.com/FlagOpen/FlagGems

enable/check flaggems functionality on triton-xpu
benchmark the performance of flaggems with triton-xpu on several OOB models
Analyze and break down the performance gap if exists
Check what the missing bits are

zhczhong added research performance labels

zhczhong self-assigned this

zhczhong mentioned this issue

[sycl-free-inference-for-llms] run llama3-8B with Pytorch for xpu and get the base line #2197

Open

vlad-penkin added this to the 4.3 [Performance] Tracking milestone

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