Roshan Karthik Rajesh comments

Repositories
Issues
Comments

Results 3 comments of


                                            Roshan Karthik Rajesh

No user-directed error if model fails to load

Hey, I am trying to run Ramalama with a RAG that's been vectorized from a URL, but when running Ramalama through ```ramalama run --rag ./uml_website/ gemma3:12b```, I get an error...

[Linux/Flatpak] Extension permission dialog cut off - Cannot install any extensions

I am facing this issue too!

rag: ramalama serve --rag doesn't take into account of RAG

It would be great if llama.cpp GUI uses the RAG proxy!