-
Notifications
You must be signed in to change notification settings - Fork 1.1k
Support CNAI OCI artifact models #9127
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
Comments
@chlins do you mind making the image |
Of course, it has already been set to public. I apologize for forgetting that the default is a private project. |
Thanks. Looking at
I think the new handling as artifacts is fine, but we do not seem to support the untar operation yet. This sounds more like an enhancement of the current implementation rather than a bug. |
@cri-o/cri-o-maintainers I assume it would be great for CRI-O to support CNAI artifact models, so that's the feature. @chlins is the image |
@saschagrunert Thanks for your confirmation, I've tested another artifact which bundled with raw file instead of tar, but also not work as normal. The raw artifact is: |
Yes, this image was packaged by modctl which followed the CNAI model spec, Although more people and projects are still involved in discussing this spec, but I believe the basic version is relatively stable. |
@chlins ah I think I'll have a fix for that! |
What happened?
It seems that the CRI-O has supported mount OCI Artifact in the main branch, but I ran into the issue when mount the artifact which bundled the AI models into it. The mount action is success, but cannot list the accurate files in pod.
What did you expect to happen?
Mount the OCI Artifact into the pod and can list the accurate files in the pod mount path.
How can we reproduce it (as minimally and precisely as possible)?
Exec into the pod and ls the
/tmp/model
.Anything else we need to know?
No response
CRI-O and Kubernetes version
OS version
Additional environment details (AWS, VirtualBox, physical, etc.)
The text was updated successfully, but these errors were encountered: