MiniMax H3 Image

Released
2026-11-10
Last refreshed
2026-11-10
Status
Researched today
Open weightsCommercial use: restrictedMultimodal
MiniMax releases · 11 in the last 12 months · this family litChangelog →
Specifications
Released
2026-11-10
Specialization
image-generation
Openness
Open weights
Weights
Available
Code
Unknown
Training
Fine-tuned
Created by
MiniMax

Developing AI for gaming and entertainment.

Minhang, Shanghai, China
Founded 2021
Website
Pricing
Output / 1M
$3.80
Input / 1M
$2.30

Cheapest of 2 routes · MiniMax

Providers(2)
MiniMaxfal API
View 2 provider routes
Links
Website HuggingFace

About

MiniMax H3 Image is MiniMax's open-weights image generation and editing model, released November 10, 2026 and built on its MiniMax H3 video model. It supports text-to-image and image editing with 1 to 16 reference images and output up to 20,000,000 pixels, with optional prompt enhancement. A faster MiniMax-H3-Image-Highspeed version takes the same parameters at a higher token price. The API is token-billed: $2.30 per 1M input text tokens, $1.50 per 1M input image tokens and $3.80 per 1M output image tokens.

Providers(2)

Compare all 2

Compare API pricing across 2 tracked provider routes for input and output tokens, batch, and cached reads when available.

ProviderInput / 1MOutput / 1MRoute
MiniMax$2.30$3.80
Serverless
fal API$2.30$3.80
Serverless

Capabilities

VisionMultimodal

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.

API versions

MiniMax-H3-ImageMiniMax-H3-Image-Highspeed