Skip to content

Asking to Release FP6 of model #8

@feyzollahi

Description

@feyzollahi

Hello
My GPU is A6000 Ada 48GB VRAM and it cannot support FP8 or AWQ-8bit varent. so I ask you to release FP6. Its accuracy is better than AWQ-4bit .
I mean this model: https://huggingface.co/moonshotai/Kimi-Linear-48B-A3B-Instruct
Thank you.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions