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.

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

Fix erroneous output in CUDA containerfile #322

Merged
merged 1 commit into from
Oct 17, 2024
Merged

Fix erroneous output in CUDA containerfile #322

merged 1 commit into from
Oct 17, 2024

Conversation

bmahabirbu
Copy link
Collaborator

Updated the containerfile to fix incorrect outputs during model execution with certain prompt sizes.
Ensured compatibility with Docker and Podman. Related to issue #9848.

@bmahabirbu
Fix erroneous output in CUDA containerfile
a9d57cd 
Updated the containerfile to fix incorrect outputs
during model execution with certain prompt sizes.
Ensured compatibility with Docker and Podman.

Signed-off-by: Brian <[email protected]>
@ericcurtin
Copy link
Collaborator

We can't push this one to quay.io like the other images due to Nvidia licensing restrictions... It's tempting to get ramalama to podman/docker build this Containerfile when it detects and Nvidia GPU... That would be obeying the rules I think...

@p5
Copy link
Contributor

p5 commented Oct 16, 2024

We can't push this one to quay.io like the other images due to Nvidia licensing restrictions... It's tempting to get ramalama to podman/docker build this Containerfile when it detects and Nvidia GPU... That would be obeying the rules I think...

Sorry to chime in, but could you elaborate why this is the case, when other projects also distribute these CUDA assets via OCI images?
Is it because RedHat is a for-profit company?

@ericcurtin
Copy link
Collaborator

We can't push this one to quay.io like the other images due to Nvidia licensing restrictions... It's tempting to get ramalama to podman/docker build this Containerfile when it detects and Nvidia GPU... That would be obeying the rules I think...

Sorry to chime in, but could you elaborate why this is the case, when other projects also distribute these CUDA assets via OCI images? Is it because RedHat is a for-profit company?

Not sure of the specifics, but all I know is we don't have clearance to do it in the RamaLama case, at least not yet, hopefully it changes in future.

@rhatdan
Copy link
Member

rhatdan commented Oct 17, 2024

LGTM

@rhatdan rhatdan merged commit 499279c into containers:main Oct 17, 2024
13 checks passed
@bmahabirbu bmahabirbu deleted the update-cuda-containerfile branch October 24, 2024 06:02
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ericcurtin ericcurtin ericcurtin approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@bmahabirbu @ericcurtin @p5 @rhatdan