Skip to main content

MiniMax H3 Max

A scene to try, with motion and sound together.

Function

Choose an image-based function above to add images, or continue with a text prompt.

Describe what you want to create.
Runs on

MiniMax H3 Max5–15 s · 1080p

Aspect ratio
Duration
Resolution
Quality
Prompt enhancement
Try a prompt:
The Look Back
The Look Back

A young adult woman in her early twenties — chestnut auburn hair, worn tan suede bomber jacket, blue jeans — walks along a rain-wet European high street lined with shop windows. She begins facing away from camera, walking at an easy unhurried pace along the glistening pavement [Tracking shot]. After a couple of natural steps she gently turns her head and then her upper body back over her left shoulder toward the friend filming her, her chestnut hair catching a little breeze as she pivots [Pan left]. She settles into a small, warm, unposed smile directed back at the camera, torso three-quarters turned, jacket catching the warm golden post-rain light that glows off the wet stone pavement and shop window glass [Static shot]. Throughout: subtle handheld camera drift, no cuts, no dramatic gestures, no model poses — purely candid and natural. Warm low-angle post-rain sunlight, golden reflections on wet pavement, slightly hazy sky. Sound: soft footsteps on wet stone, quiet ordinary street ambience, distant city murmur, light breeze; no speech, no music, no captions, no watermark, no logo.

MiniMax H3 Max · 5 s

Quiet location audio accompanies this street portrait motion study.

How it works

Try it now
  1. 01

    Say what you want

    A sentence or two is enough.

  2. 02

    Use MiniMax H3 Max

    Choose the settings that fit what you want to make.

  3. The Look Back
    03

    Make it

    Start creating, then keep working with HowWorks.

Explore the possibilities

The Look Back
Quiet location audio accompanies this street portrait motion study.MiniMax H3 Max

MiniMax H3 Max at a glance

Good at
Fastest to render, with a Turbo tier for quick drafts. 5–15s, up to 1080p.
Modes
Text to Video, Image to Video, First & Last Frame, Reference to Video
Length
5–15 s (presets 5, 10, 15 s)
Resolution
480p, 720p, 1080p
Aspect ratios
16:9, 9:16, 1:1, 4:3, 3:4, 21:9
Image-led framing
May follow the input image; check the output frame
Sound
Always on
Reference images
Up to 9
Tiers
Standard · Turbo

About

Keep MiniMax H3 Max selected while you explore a short scene from text or an image. For creators writing an audible action or a quiet atmosphere, describe the picture and the sound together: rain against a window, a cup touching a table, a room of chimes. Audio is included with the video. Choose from the output settings shown for your request, then watch and listen to the actual result before using it in a sequence or social post.

Questions

Can I switch the generated audio off?

H3 Max’s current model configuration includes audio automatically. It does not use the Seedance audio on/off setting. Describe the sound you want, then check what is actually audible.

Can I rely on exact speech or sound timing?

Review the actual audio rather than treating the prompt as a transcript. If exact wording or timing is essential, check every line and cue before using the clip.

Can I animate an image with H3 Max?

Yes. Add a start image in Image to Video and describe the movement. H3 Max follows the source image’s aspect in this mode, so choose its composition before uploading.

Does H3 Max reach the same output resolution as H3?

H3 Max’s model configuration reaches 1080p; H3 has a different resolution range. Check the options shown for your request instead of assuming that Max means a larger output.

What changes between Standard and Turbo?

The model configuration distinguishes their supported inputs: Turbo does not support reference-image generation. Use the choices actually shown for your request; a tier listed for the model does not mean every input workflow is available in this website entry.