> For the complete documentation index, see [llms.txt](https://docs.pletor.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pletor.ai/models/image-models/minimax-h3-image.md).

# MiniMax H3 Image

MiniMax's image model for generation and editing: up to 16 references in one edit, native 4K, and a prompt expander that turns short briefs into detailed descriptions.

## TL;DR

The image sibling of [Minimax H3](https://docs.pletor.ai/models/video-models/minimax-h3.md), taking more references than [Seedream 5.0 Pro](https://docs.pletor.ai/models/image-models/seedream-5.0-pro.md) or [Flux 3 Image](https://docs.pletor.ai/models/image-models/flux-3-image.md) (16 vs 10). Two versions with the same settings: **MiniMax H3 Image** (default) and **Highspeed** (faster, higher cost per image).

### Strengths for marketers

* **Multi-reference composition**: up to 16 images into one shot.
* **Short prompts work**: every prompt is expanded into a full description.
* **Native 4K**: 1K, 2K (default) or 4K.
* **Flexible framing**: 1:3 to 3:1, including 2.35:1 and 9:16.

### Ideal use cases

* Bundle and set shots from several SKU photos.
* On-model product shots.
* Fast concepting from one-line briefs.
* 4K key visuals.

### Weaknesses

* **Your wording gets rewritten**: put must-appear copy in quotes and check it.
* **Auto is square from text** (2048×2048); when editing, Auto keeps the first image's shape.
* **No masks or transparency**.
* **Refer to references by number**, in connection order.
