[ci] uniform run times and wheels for pytorch cuda. #13047
+6
−4
Conversation
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
DN6
reviewed
Jan 29, 2026
| @@ -1,4 +1,4 @@ | |||
| FROM nvidia/cuda:12.9.0-runtime-ubuntu20.04 | |||
| FROM nvidia/cuda:12.9.1-runtime-ubuntu22.04 | |||
Collaborator
There was a problem hiding this comment.
Is there no wheel for ubuntu24.04?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
What does this PR do?
Sadly the previous PR didn't quite fix it. I did locally build with the updated Dockerfiles on A100 and ran some tests, but nothing like this showed up there. I even tried to completely recreate the run as much as possible and still, nothing showed up. I guess it's one of those scenarios where things show up when you run them in the full roll.
The issue is that even when we specify CUDA 12.9, PyTorch fails to find any wheel for it because we only have 12.8 and 13.0. It was installing the 12.8 wheel and that is causing a mismatch in the CUDA runtime currently.Edit: Changed to a different base image and specified thr 129 CUDA wheel as it was available.