Primary use stated or inferred cautiously from official documentation.
MiniMax-H3-Image
MiniMax image model derived from H3, with weights published on November 10. It accepts text instructions and visual references for image generation and composition editing.
The essential
Maximum input capacity when published by the source.
Documented maximum generation limit.
Verified availability channels.
Primary functional family and verified specialties.
Declared terms for API access, model weights, or self-hosting.
Limits and integration
| API ID | MiniMax-H3-Image |
|---|---|
| Model type | Image |
| Access model | Downloadable weights and API |
| License | MiniMax-H3-Image Community License Agreement |
| Deployment | API and self-hosting |
| Release | 10 NOV 2026 |
| Knowledge cutoff | Not published |
| Entrance | Text · Image |
| Exit | Image |
| Context window | 32,000 characters and up to 16 images |
| maximum output | 1 to 10 images, up to 20 million pixels per image |
| Reasoning | Not published |
| Published tools | |
| Structured Outings | Not published |
| Batch processing | Not published |
| Prompt cache | Not published |
| Fine-tuning | Not published |
| Verified platforms | MiniMax API · Hugging Face · Diffusers · ComfyUI |
Create illustrations, product concepts and visual variations from up to 16 references. The API offers PNG, JPEG and WebP. Output aspect ratios range from 1:3 to 3:1.
Using the full 32,000-character limit requires prompt expansion. With expansion disabled, only about 4,096 tokens take effect. The API does not support transparent backgrounds or mask-based editing.
What it can do
Orientation
Create illustrations, product concepts and visual variations from up to 16 references. The API offers PNG, JPEG and WebP. Output aspect ratios range from 1:3 to 3:1.
Context
32,000 characters and up to 16 images · 1 to 10 images, up to 20 million pixels per image
Tools and integration
Not published
Access
MiniMax API · Hugging Face
Documented cost
| Concept | Worth | Unit/condition |
|---|---|---|
| Standard input | Text $2.30 · image $1.50 / 1M tokens | Standard API rate |
| Cached input | Not published | Reading reused prefixes |
| Cache write or storage | Not published | The condition varies by provider |
| Standard output | Image $3.80 / 1M tokens | May include reasoning tokens |
| Batch input | Not published | Asynchronous processing |
| Batch output | Not published | Asynchronous processing |
Per million tokens, Standard costs $2.30 for text input, $1.50 for image input and $3.80 for image output. Cost depends on resolution and the expanded text.
i Prices change and may depend on level, region or context length. Check the source before making a decision.
How to read the results
The announcement compares speed, instruction following and visual quality. Manufacturer measurements depend on deployment and configuration. Inferama has not reproduced these tests.
i Manufacturer comparisons in the launch announcement.
Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.
Chronology
MiniMax-H3-Image
Image generation and editing