# MiniMax releases H3-Image, an open-weight 16-megapixel image generation and editing model

> Original title: MiniMaxAI/MiniMax-H3-Image

- Source: MiniMax · new models on Hugging Face
- Published: 2026-11-03T06:34:36.000Z
- Bad Influence: https://badinfluence.news/items/e7k83rvtzd45abdcgc6q2v5sx
- Original: https://huggingface.co/MiniMaxAI/MiniMax-H3-Image

## Summary

MiniMax has released MiniMax-H3-Image as an open-source model for text-to-image generation and image editing, with weights on Hugging Face. The model generates about 16-megapixel images at every aspect ratio, from 4096x4096 to 6272x2688, using 6 sampling steps without classifier-free guidance. It supports up to 16 reference images, and the repository lists deployment options including SGLang, vLLM-Omni, diffusers, and ComfyUI.

## Why it matters

The release documents a 16-megapixel open-weight image model with setup and sizing details, useful for judging whether local deployment fits a given workflow.