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

[helm-charts] TGI parameters for Gaudi #499

Closed
yongfengdu opened this issue Oct 25, 2024 · 0 comments · Fixed by #528
Closed

[helm-charts] TGI parameters for Gaudi #499

yongfengdu opened this issue Oct 25, 2024 · 0 comments · Fixed by #528
Labels

Comments

@yongfengdu
Copy link
Collaborator

Current helm chart for TGI doesn't have much Gaudi specific parameters setting like:
OMPI_MCA_btl_vader_single_copy_mechanism: none
ENABLE_HPU_GRAPH: true
LIMIT_HPU_GRAPH: true
USE_FLASH_ATTENTION: true
FLASH_ATTENTION_RECOMPUTE: true
(https://github.com/opea-project/GenAIExamples/blob/main/VisualQnA/docker_compose/intel/hpu/gaudi/compose.yaml)
We need to evaluate the impact of the tei-gaudi parameters and have the optimized settings.
This could be part of a bigger scope of tuning/optimization.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants