Update references to Hugging Face DLC for TGI #816
Conversation
alvarobartt
requested review from
achandrasekar,
ahg-g and
annapendleton
as code owners
...nference-server/text-generation-inference/manifest-templates/text-generation-inference.tftpl
Outdated
Show resolved
Hide resolved
Included a harmless TODO note so that we remember to come back to this once the benchmark is verified with the latest Hugging Face DLC for TGI cc @annapendleton
|
/gcbrun |
|
@alvarobartt I also added you as a collaborator so you can test + merge this yourself :) |
|
Hi here @annapendleton, so the CI is failing, but not sure that's something I can fix or related to any of the changes within this PR, could you help me debug the issue further if related to the PR? 🤗 |
From what I can see in the logs, it looks like the test cluster isn't present. We don't use these tests ourselves for the benchmarking, I think we should add someone else who has a better sense of if it's safe to bypass this test or not. |
|
@imreddy13 could you help us determine if this is a true test failure? |
Description
This PR updates a bunch of outdated references to Text Generation Inference (TGI) container, as previously those were pointing to the GitHub Container Registry where Hugging Face publishes those containers i.e. ghcr.io/huggingface/text-generation-inference; but since the Hugging Face DLCs on Google Cloud were recently released publicly (see
Google-Cloud-Containers), now those references have been updated to point to the DLCs hosted on Google Cloud's Artifact Registry.cc @brandonroyal