MiniMax H3: Best Practices

Start With the Whole Scene

A useful H3 prompt describes the creative result, not a list of disconnected keywords.

Instead of: Woman, train, cinematic, rain, camera movement.

Try: A cinematic close shot of a woman sitting beside the window of a moving train at night. Rain runs across the glass as city lights streak through her reflection. She looks down at a folded letter, then slowly raises her eyes toward the window. The camera gently tracks to the right. Train wheels rumble underneath quiet rainfall, with sparse piano playing in the background.

The second version gives H3 relationships to work with: subject, environment, action, camera, timing, ambience, and music, rather than isolated terms.

Choose the Right Starting Point

H3 can work from several kinds of creative context, depending on what you provide:

• Text only: describe the complete scene from scratch.

• A starting image: treat it as the opening state, and describe what begins moving and how the scene develops from there.

• Starting and ending images: focus on the path between them: the physical changes, camera movement, subject motion, and transitions that lead naturally from the first composition to the last.

• References: different assets can each contribute a different part of the result. An image might define the character, a video might define camera movement, and audio might define a voice or musical direction. MiniMax calls this its Omni Reference workflow.

Think Beyond the Picture

H3 generates audio together with video, so sound can be part of the original creative direction instead of an afterthought. Consider describing:

• Dialogue: Without looking away from the window, she quietly says, "I think this is my stop."

• Environmental audio: Rain taps against the glass while the train produces a steady metallic rhythm.

• Physical sounds: The paper softly unfolds in her hands.

• Music: Sparse piano notes play at a slow tempo, joined by a sustained low cello.

MiniMax's own prompting guidance separates these into three layers: the visual and action timeline, the overall soundscape, and non-diegetic music. Not every prompt needs all three, the useful habit is simply thinking about the complete audiovisual moment rather than just the picture.

Give Every Reference a Clear Job

More inputs don't automatically make a generation better. If you're combining several references, be explicit about what each one is for, for example, an image defines the character, a video defines motion, and audio defines the voice. Clear relationships give the model a stronger creative hierarchy than a pile of unexplained references.

The Core Habit

Describe what should happen, how it should look, how it should move, and what the audience should hear, and when using references, say what each one contributes to the final result.