AI Video APIAI Video API

MiniMax H3 Image

Text-to-image, image editing and independent prompt enhancement.

All models

Platform safety check, enabled by default. Disabling requires account permission; otherwise it stays enabled.

On

Up to 32,000 Unicode characters. Built-in prompt enhancement is billed as input text tokens.

Custom dimensions use 655,360–20,000,000 pixels and ratios from 1:3 to 3:1. Custom size replaces area and aspect_ratio.

Text-to-image only. Disabling enhancement requires a concrete size or ratio.

Off
Off
Pricing
Billed on completion
Actual token usage

Rates per 1 million tokens, in CNY:

Text input¥19.50
Image input¥13.00
Image output¥32.50

No pre-charge. Submit with a positive balance; successful work is billed once using actual usage. The final debit may make your balance negative. Top up before starting another task.

Credits use 6.67 CNY/USD and USD 0.005/credit, rounded to two decimal places.

4K offer: 50% off the whole request when every successful image is at least 16.6 million pixels. Dispatch must be before November 18, 2026, 00:00 Beijing time. Mixed or missing sizes require billing review.

Result
View Logs

Configure parameters and click Run to test the API.

Code Example
curl -X POST https://aivideoapi.ai/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "minimax-h3-image",
  "input": {
    "nsfw_checker": true,
    "size": "auto",
    "aspect_ratio": "auto",
    "area": "4K",
    "n": 1,
    "disable_prompt_enhanced": false,
    "output_format": "jpeg",
    "output_compression": 95,
    "aigc_watermark": false
  }
}'