Skip to content

[feature request] LLaVA-1.6 update #240

Open
@shigabeev

Description

New model seems to significantly improve upon v.1.5 which is the default one used in README.md. I'm eager to update my llamafile with this model.

I started reading README on making my own llamafile and it seems that I need to first convert my model into gguf and from then on verything seems to be convered in README.md. But what are the steps before? Should I quantize model myself or is it done on the side of this repo?
Can you walk me which steps should I undergo to convert the model weights from HF into the llamafile?

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Assignees

No one assigned

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions