Ian Eaves
Ian Eaves
@dependabot recreate
@ericcurtin `ramalama run granite3-moe` is working for me on 0.8.3 / after merging #1367. This still fails when called as bin/ramalama though. Is there any need to fix the bin...
If the issue is related to #1367 then the issue would be environment specific. In my case, /bin is still polluting my path leading python to attempt importing libraries from...
The problem was already solved in #1367 for anyone calling `ramalama` normally (it's fixed by `project.scripts` in the pyproject.toml). It's only an issue for someone manually invoking /bin/ramalama. I think...
I think we might be missing each other. Don't worry about it, #1340 fixes the issue for anyone installing with `pip install ramalama`.
I _believe_ this was partly addressed in the vision PR for chat but our chat interface assumes an openai API compatible server so: > The primary issue is neither ollama...
I believe so! Should be good to close.
@dependabot recreate
@dependabot recreate
@dependabot rebase