Add examples/gke/tgi-multi-gpu-deployment/ for multi-GPU TGI#58
Draft
alvarobartt wants to merge 7 commits intomainfrom
Draft
Add examples/gke/tgi-multi-gpu-deployment/ for multi-GPU TGI#58alvarobartt wants to merge 7 commits intomainfrom
examples/gke/tgi-multi-gpu-deployment/ for multi-GPU TGI#58alvarobartt wants to merge 7 commits intomainfrom
Conversation
Member
Author
|
Note Since the TPU containers are still WIP and not released yet on Google Cloud, the examples are on hold until those are released; meaning that most likely the current example would need to be revisited before merging. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
This PR adds an example on how to serve
google/gemma-7b-itin Google Kubernetes Engine (GKE) using 4 x L4 GPUs, which better emulates a production environment, as well as explaining some concepts in order to have a production-ready deployment.