Export
The link always opens the latest version of this page.
Ask an AI about it
Open in ChatGPTA new chat that reads this entry from the site Open in ClaudeThe same, in ClaudeShare the link instead of a file, and they will always see the latest version.
Files are a snapshot of today. The live page has anything newer.
MiniMax-H3-Image
Unified text-to-image and image-editing diffusion model supporting high-resolution generation up to 16 megapixels
MiniMax-H3-Image is a unified text-to-image and image-editing diffusion-model developed by MiniMaxAI 1.
The model operates as a diffusion transformer architecture and reportedly uses Qwen3-VL as its text encoder 1. It supports high-resolution image generation up to 16 megapixels.
According to available documentation, the model fixes inference steps at 6 and requires guidance scale to be set to 1 or omitted entirely 1.
Facts 2
| Attribute | Value | Confidence | Sources | Cited |
|---|---|---|---|---|
| Vendor | MiniMaxAI | low | 1 | 1 |
| Category | text-to-image and image-editing model | low | 1 | 1 |
Version 1
- Inference steps fixed at 6; guidance scale must be 1 or omitted