Daniel J Walsh

icon indicating a website link http://danwalsh.livejournal.com icon indicating an email [email protected]

icon company Red Hat icon location Wells, ME icon location Mr SELinux, Red Hat Senior Distinguished Engineer. Works on @containers projects: ai-lab-recipes, Podman, Skopeo, Buildah, image, storage...

Results 1685 comments of Daniel J Walsh

Using vllm runtime generates "unrecognized arguments" error

Can you try again with 0.12.1? Usually the mismatch is caused by a screwed up cuda library install.

Using vllm runtime generates "unrecognized arguments" error

Grab 0.12.2 and try serve to get the error the service is hitting.

Cannot run converted models (imported using convert command)

Can you run the ollama model successfully?

Cannot run converted models (imported using convert command)

``` ollama run pllum8b-instr-q4km pulling manifest Error: pull model manifest: file does not exist ```

Cannot run converted models (imported using convert command)

Are you still seeing this issue?

Cannot run converted models (imported using convert command)

@dwrobel Still seeing this issue?

Cannot run converted models (imported using convert command)

Thanks, hope to take a look at this one.

Cannot run converted models (imported using convert command)

@engelmi The model_store has broken this support. Basically we need to mount the image not the path when using OCI images.

Cannot run converted models (imported using convert command)

We close issues, when they are closed in the upstream, not in the latest release. @engelmi did you discover if this was not fixed in the main branch?

Better chat template handling - support Jinja

@edmcman @ericcurtin any update on this?