Skip to content

Support MiniMax-H3 Visual VAE quantization in ModelOpt #2137

Description

@baonudesifeizhai

vLLM-Omni is currently working on production VAE optimization for MiniMax-H3 in
vllm-project/vllm-omni#5948

As a follow-up, would the ModelOpt team be interested in adding quantization support for the MiniMax-H3 Visual VAE?

This could explore FP8, NVFP4, or mixed-precision quantization depending on quality and kernel support, and potentially integrate with the VAE optimization path in vLLM-Omni.

Thanks!

Activity

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

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions