Text-to-video and first/last-frame video API

MiniMax H3 Max Turbo API

Create 480p or 768p video from text and controlled frames.

MiniMax H3 Max Turbo generates 5–15 second clips from prompts or one to two frame images. Choose prompt expansion and an optional seed, then collect the video through the unified task API. Credits depend on requested output seconds and resolution; input images and prompt expansion are free.

Try playground

5–15s

Output duration

480p / 768p

Resolution

6.54 credits/s

Starting price

MiniMax

minimax-h3-max-turboH3 family sample

5–15s

Output duration

480p / 768p

Resolution

6.54 credits/s

Starting price

Model capabilities

Built for production API workflows, not one-off demos.

Use MiniMax H3 Max Turbo from the same platform surface as the rest of your video and image stack: API keys, credits, logs, webhooks, and docs stay consistent across providers.

Text-to-video

Omit generation_type to generate from a prompt. Choose 21:9, 16:9, 4:3, 1:1, 3:4, or 9:16; the default is 16:9.

First and last frame control

Use first_and_last_frames with one public image URL for the opening frame or two for both endpoints. The output aspect ratio adapts to the images.

Prompt expansion and seed

Choose disabled, balanced, or quality prompt expansion; balanced is the default. Optionally set an integer seed from 1 through 4294967295.

Output-only pricing

480p costs 6.54 credits per requested second and 768p costs 10. Input images and prompt expansion never add fees; your account multiplier applies to the base charge.

API workflow

Submit tasks, track progress, and return generated assets.

Choose your generation mode

Use a prompt of up to 7,000 characters. Omit generation_type for text-to-video, or set first_and_last_frames and provide one or two public HTTP(S) image URLs.

Set the output controls

Choose an integer duration from 5 to 15 seconds, 480p or 768p, and a prompt expansion mode. Defaults are 5 seconds, 768p, and balanced expansion.

Estimate, submit, and retrieve

Use POST /v1/estimate to preview the charge, then POST /v1/videos/generations to create a task. Poll the task ID or receive a callback and collect the video from output.urls.

POST /v1/videos/generations

curl -X POST https://api.aivideoapi.ai/v1/videos/generations \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "minimax-h3-max-turbo",
    "input": {
      "prompt": "A cinematic tracking shot through a rain-lit night market",
      "resolution": "768p",
      "duration": 5,
      "aspect_ratio": "16:9",
      "prompt_expansion_mode": "balanced",
      "seed": 123456
    }
  }'

What teams build

Social video variations

Product motion previews

Storyboard transitions

Short campaign videos

Opening and ending frame control

Prompt and composition experiments

Frequently asked questions

Answers about the MiniMax H3 Max Turbo API.

Which inputs are supported?

Use text or first_and_last_frames with one or two HTTP(S) images. Text mode ignores image_urls. Multimodal references, video input, audio input, 2K output, and watermark controls are not supported.

Which durations can I submit?

Only integers from 5 through 15 seconds are accepted. The default is 5 seconds. Invalid durations are rejected before credits are charged.

How are credits rounded?

Multiply requested output seconds by the resolution rate and round the full base charge to two decimal places, apply your account multiplier, then round again. For 480p, 7 seconds, and a 1.1 multiplier, the charge is 50.36 credits. Returned media usage does not add fees.

What happens if generation or delivery fails?

A confirmed failed task is refunded once. An uncertain submission keeps its pre-charge for reconciliation. A successful generation awaiting video delivery keeps its pre-charge while delivery retries. Submitted tasks cannot be cancelled.

Pricing and usage

Clear model options with shared credits.

480p output

6.54 credits/s

5s: 32.70 credits; 15s: 98.10 credits before account multipliers

768p output

10 credits/s

Default resolution; 5s: 50 credits; 15s: 150 credits before account multipliers

Input images and prompt expansion

Free

Text, first-frame, and first/last-frame generation use the same output rates

Start building with MiniMax H3 Max Turbo in AI Video API.

Create one API key, use one credit balance, and switch between video and image models without provider-specific plumbing.

Read docs