MiniMax H3 Max Turbo brings prompt-faithful video generation to text and frame workflows at a lower credit cost than H3 Max.

What does the output look like?

Animated typography

Animated lettering with generated sound from fal’s published H3 Max Turbo demonstration.

Provider demonstration: fal H3 Max Turbo demonstration

What can MiniMax H3 Max Turbo do?

MiniMax H3 Max Turbo video generation from text or first and last frames, with clips up to 15 seconds and 1080p refinement

  • Lower-cost video iteration from text
  • Animating a start image
  • Controlling a transition with start and end frames
Inputs
Text, Image
Durations
5s, 6s, 7s, 8s, 9s, 10s, 11s, 12s, 13s, 14s, 15s
Aspect ratios
21:9, 16:9, 4:3, 1:1, 3:4, 9:16
Output options
480p, 768p, 1080p
Typical generation time
About 10 minutes
  • 1080p uses latent refinement from a native 768p source. Frame generation follows the start image's canvas, or the end image when only an end frame is provided.
  • Turbo accepts text and frames; choose H3 Max for mixed image, video, and audio references.

How much does MiniMax H3 Max Turbo cost?

Base estimate

$0.025–$0.08 USD / second

0.25–0.8 credits / second

Estimated dollar cost uses the Solo monthly plan at $0.10 USD per credit. Your cost per credit depends on your plan.

Example: a 5-second clip

$0.125–$0.40 USD(1.25–4 credits)

Reference media and video editing may add to this estimate.

Video cost scales with duration. Resolution, sound, reference media, and generation mode can affect the final charge. Check the total shown in Starpop before generating.

How to use MiniMax H3 Max Turbo

  1. Click “Create with MiniMax H3 Max Turbo” and sign in or create an account. After setup, Starpop opens a new video workspace tab with the model selected.
  2. Describe a scene or attach first and last frames for a controlled transition.
  3. Choose the duration and resolution. Use a different route if the shot needs uploaded audio or video references.
  4. Review the credit estimate and generate. Check the result in your workspace, then download it or adjust your inputs to try again.

More about using this model