Overview
What to know before using H3
MiniMax H3 is a multimodal AI video model built for video generation with text, image, video, and audio context. It can be used for short videos, product ads, cinematic scenes, character animation, app promos, brand films, and reference-based video creation.
The important idea for users is simple: a good MiniMax H3 workflow is not only a text prompt. It is a video brief that explains the subject, scene, action, camera, audio, references, constraints, and output settings.
Before You Start
What to prepare before using H3
Video goal
Decide if the output is a product reveal, short video, cinematic scene, character shot, or app promo.
Creative direction
Prepare the subject, scene, camera movement, visual style, audio direction, and negative constraints.
Reference materials
Use references when you need stronger control over subject identity, movement, composition, or sound.
Step-by-step Tutorial
Step-by-step H3 video workflow
Choose your video goal
Start with one clear job: product ad, cinematic scene, short social video, character animation, or app promo. This keeps the brief focused and prevents the prompt from mixing unrelated goals.
Describe the subject
Write what must stay consistent: product shape, character appearance, material, color, logo placement, pose, clothing, or key visual details.
Define the scene
Describe where the action happens. Keep the scene separate from reference instructions so references can guide subject, motion, style, audio, or composition as needed.
Add camera movement
Use specific direction such as slow dolly-in, overhead shot, macro close-up, handheld tracking, crane-in, wide establishing shot, or final hero frame.
Add audio direction
According to MiniMax, H3 supports native audio generation. Describe music, ambience, sound effects, dialogue tone, rhythm, and mood. so describe music, ambience, sound effects, dialogue tone, rhythm, and the desired emotional energy.
Add reference instructions
Explain how @image1, @video1, or @audio1 should influence the output. Do not assume the model knows which file controls which creative decision.
Choose settings
Select duration, resolution, aspect ratio, and output purpose before generation. Keep short videos simple and reserve longer clips for multi-shot scenes.
Review and refine
Check the generated brief, shots, reference plan, negative prompt, and cost estimate. Refine before the final video task to avoid wasted generations.
Video Brief Formula
How to write an H3 video brief
A strong MiniMax H3 brief should explain what the model should generate and how each reference should guide the result.
Subject + Scene + Action + Camera Movement + Visual Style + Audio Direction + Reference Instructions + Duration + Resolution + Constraints
Example video brief
Create a 10-second 2K MiniMax H3 product video for a black smart watch. Subject: A black smart watch with a glowing digital display and premium metal body. Scene: Night city street with wet reflective pavement, neon signs, and a premium technology atmosphere. Camera: Slow dolly-in close-up with shallow depth of field. Audio: Deep electronic music, subtle city ambience, and soft interface sound effects. Reference instructions: @image1 should guide the watch shape, color, material, display design, and key product details. @audio1 should guide the mood, tempo, ambience, and sound texture. Constraints: Avoid distorted product shape, unreadable text, extra logos, unrealistic reflections, and shaky camera.
Multimodal References
How to use image, video, and audio references in an H3 workflow
References should not be hidden inside the scene description. Treat them as a separate instruction layer. A reference can guide the subject, scene composition, motion, camera rhythm, style, or audio.
@image1
Use image references for subject identity, product shape, color, texture, face, outfit, layout, composition, or scene details.
@video1
Use video references for action timing, motion style, camera rhythm, character movement, scene continuity, and transitions.
@audio1
Use audio references for music style, ambience, tempo, emotional tone, voice direction, sound effects, and audio-video rhythm.
Reference instruction examples
@image1 should guide the subject appearance, product shape, color, texture, and key visual details. @video1 should guide camera rhythm, motion pacing, character action, and scene continuity. @audio1 should guide mood, tempo, ambience, voice tone, and sound texture.
Workflow Examples
H3 prompt examples by workflow
Use these examples to understand how different H3 workflows change the way you write a video brief. Each example separates the input type, reference role, camera direction, motion control, style, and constraints.
Vintage binocular brand film
Use this pattern when multiple images should act as ordered keyframes and the camera language matters as much as the subject.
Use Images 1-4 as sequential keyframes, seen through a vintage binocular viewfinder searching for the MINIMAX installation. Open out of focus with subtle handheld shake, then push in quickly and rack focus onto Image 1. Between keyframes, use fast binocular-scan transitions with whip movement, motion blur, optical smearing, and brief exposure flicker. Keep the twin circular lens mask absolutely fixed throughout. Visual language: a voyeuristic, Wes Anderson-inspired 35mm film look with fine grain, soft highlight halation, restrained color, and red typographic accents.
Fantasy wuxia character film
Use this pattern when one image locks the character design and another reference controls storyboard order or pacing.
Use Image 2 as the locked character reference. Preserve the half-up long black hair, openwork silver crown, indigo ribbon, layered pale hanfu, translucent blue outer robe, deep-blue sash, silver floral fastener, and long tassels. Use Image 1 for storyboard order and pacing. Render in high-quality 2K, 16:9 Chinese-inspired 3D with cinematic xianxia production value: intense, solemn, and shaped by destiny. Follow the storyboard beat by beat, with natural camera movement and seamless transitions.
Live-action to voxel transformation
Use this pattern when the source video should preserve the real environment while selected elements change style.
Preserve the buildings, pedestrians, and overall environment in Video 1 as photoreal live action. Transform only the trees and cars into 3D pixel-art or voxel-block objects in the style of Minecraft, using Image 1 as the visual reference. Keep their motion physically correct, and preserve the real environment's shadows and transmitted light. Use Video 2 as the overall target.
Neon laundromat encounter
Use this pattern when there are no uploaded references and the text must fully define scene, camera behavior, mood, and authenticity.
15 seconds, 16:9 landscape. Combine a live-action late-night laundromat with hand-drawn luminous animation. The small self-service laundromat has gently flickering fluorescent lights, running washers, plastic baskets, a worn bench, and one sock on the floor. Keep the space quiet and faintly nostalgic. Use a one-handed phone-camera feel with visible shake, exposure fluctuation under white fluorescent light, environmental reflections in glass, and delayed autofocus at close range. Avoid polished commercial composition; it should feel like an authentic late-night encounter, filmed while following a strange apparition.
Motion path recreation
Use this pattern when a reference video controls exact movement, timing, camera lock, and spatial choreography.
Match the action in Video 1 from a locked-off wide camera. Replace the three suited men with three highly photoreal capybaras. Preserve the original movement path exactly: all three drop quickly to the floor; the left capybara jumps to center; the center capybara rolls to the far left; the new center capybara rolls to the far right; the right capybara jumps to center; finally, the center capybara jumps onto the other two, forming a pyramid. Keep the camera fixed and integrate fur, lighting, and shadows realistically into the scene.
Green-screen fairytale composite
Use this pattern when the task is compositing: preserve the subject performance while replacing and matching the background.
Remove the green screen background of Video 1 and turn it into a fairy tale-like background similar to Video 2. The background elements need to completely match the actions of the characters in Video 1. Modify the lighting of the characters in Video 1 so that it completely matches the background.
Settings
H3 settings: duration, resolution, ratio, and cost
Duration
Use shorter durations for simple single-action scenes and longer durations for multi-shot scenes that need setup, action, and final hero framing.
Resolution
Use 768P for fast drafts and planning. Use 2K when the final output needs higher detail, product clarity, or polished presentation.
Cost planning
Estimate cost review the brief before creating the video, especially when using 2K output, longer durations, multiple iterations, or batch production.
Troubleshooting
Common H3 workflow mistakes and how to fix them
Prompt is too vague
Fix it by adding subject, scene, camera, audio, duration, and specific constraints.
References are unclear
Fix it by explaining whether each reference controls appearance, motion, style, scene, or audio.
Too many goals in one clip
Fix it by reducing the video to one clear creative job and using shots to sequence the idea.
Try it online
Turn this tutorial into an H3 video brief
Use the homepage generator to create a structured brief with prompt, shots, audio direction, reference instructions, materials, and cost estimate.
FAQ
MiniMax H3 tutorial FAQ
How do I use MiniMax H3 step by step?
Start with a clear video goal, prepare the subject and scene, add camera and audio direction, define reference instructions, choose settings, then review the generated H3 video brief review the brief before creating the video.
What should I prepare before using MiniMax H3?
Prepare the video goal, subject details, scene description, camera movement, audio direction, output settings, negative constraints, and any image, video, or audio references.
How do I write a video brief for H3?
Write an H3 video brief with subject, scene, action, camera movement, visual style, audio direction, reference instructions, duration, resolution, and constraints.
How should I use image references in an H3 workflow?
Use image references to guide subject identity, product shape, color, material, face, outfit, composition, or scene details. Explain the role of each image reference in the brief.
How should I use video references in an H3 workflow?
Use video references to guide motion style, action timing, camera rhythm, scene continuity, character movement, and transitions.
How should I use audio references in an H3 workflow?
Use audio references to guide music style, ambience, tempo, emotional tone, voice direction, sound texture, and audio-video rhythm.
What are common mistakes when writing an H3 video brief?
Common mistakes include vague subjects, unclear reference roles, too many goals in one clip, missing camera direction, weak audio direction, and no negative constraints.
Where can I create a video after reading this tutorial?
You can try the H3 video generator on the homepage, where the tool helps create a structured video brief with prompt, shots, references, audio direction, materials, and cost estimate.