Skip to content

I have created an 8-bit and 2-bit quantised version but it keep loading original #55

@220940947

Description

@220940947

Due diligence

Topic

The paper

Question

I have created a 8-bit and 2-bit version but it keeps loading the original anyone else having this problem or is there a known solutions?

I have adjusted server.py so it should but it keeps falling back to the original model and not my smaller ones for better usage on more pc's.

Like it is not even looking for the model when starting the server. Even tho the path's are correct.

Metadata

Metadata

Assignees

No one assigned

    Labels

    questionFurther information is requested

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions