Skip to content

[Feature] Add ability to point to trusted external GPU / AI model / weightlift server #96

Description

@bitsondatadev

I don't enjoy melting the thing I do work on every day (a humble Framework with with 2GB vram gpu to handle simple graphics) and so I have a dual 3090 GPU server at my house to do the AI stuff for home/work.

  1. Would running weightlift on another local server work with the existing architecture, while my terminal desktop stays cool and looks pretty?
    1.1) Would weighlift already do this today as is?
    1.2) Would you consider incorporating other popular local ai frameworks such as localai/ollama?

  2. Would it interest you to have someone else work on a feature to point rescript to that server? I'm willing to roll my sleeves up and contribute as I have other features I need and have in mind, but don't want to dump time until @wassgha and friends agree that the feature is valuable + design aligns with theirs.

Thanks all

Activity

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

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions