M
MINIMAX / MiniMax-H3-Image

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.

01 / SUMMARY

The essential

ORIENTATIONImage generation and editing

Primary use stated or inferred cautiously from official documentation.

CONTEXT32,000 characters and up to 16 images

Maximum input capacity when published by the source.

EXIT1 to 10 images, up to 20 million pixels per image

Documented maximum generation limit.

ACCESSMiniMax API · Hugging Face

Verified availability channels.

MODEL TYPEImage

Primary functional family and verified specialties.

LICENSEMiniMax-H3-Image Community License Agreement

Declared terms for API access, model weights, or self-hosting.

02 / TECHNICAL SHEET

Limits and integration

API IDMiniMax-H3-Image
Model typeImage
Access modelDownloadable weights and API
LicenseMiniMax-H3-Image Community License Agreement
DeploymentAPI and self-hosting
Release10 NOV 2026
Knowledge cutoffNot published
EntranceText · Image
ExitImage
Context window32,000 characters and up to 16 images
maximum output1 to 10 images, up to 20 million pixels per image
ReasoningNot published
Published tools
Structured OutingsNot published
Batch processingNot published
Prompt cacheNot published
Fine-tuningNot published
Verified platformsMiniMax API · Hugging Face · Diffusers · ComfyUI
BEST SUITED FOR

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.

WORTH MONITORING

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.

03 / CAPABILITIES

What it can do

01

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.

02

Context

32,000 characters and up to 16 images · 1 to 10 images, up to 20 million pixels per image

03

Tools and integration

Not published

04

Access

MiniMax API · Hugging Face

Input modalities
TextImage
04 / PRICES

Documented cost

ConceptWorthUnit/condition
Standard inputText $2.30 · image $1.50 / 1M tokensStandard API rate
Cached inputNot publishedReading reused prefixes
Cache write or storageNot publishedThe condition varies by provider
Standard outputImage $3.80 / 1M tokensMay include reasoning tokens
Batch inputNot publishedAsynchronous processing
Batch outputNot publishedAsynchronous 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.

05 / EVALUATIONS

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.

06 / VERSIONS

Chronology

MiniMax-H3-Image

Image generation and editing

07 / SOURCES

Traceability