Skip to content
PromptMotion

MiniMax H3Topic · Video

MiniMax H3 Prompts

MiniMax H3 prompts from public video posts, paired with the original sources and marked translations. MiniMax describes H3 as an open, general-purpose multimodal video model for video generation, reference-based creation and video editing.

Cases7
With prompt7
LatestOct 14, 2026

01Results

MiniMax H3 Prompt Examples

MiniMax H3X post · plays on click
Video0:15Watch the original
MiniMax H3Prompt

A reference-led character detail montage

@aimikoda · 3,286 likes

Use @[character reference] for the character's identity, anatomy, costume, colors, materials and visual style. Adapt body-region cues to its actual anatomy, replacing absent features with existing equivalents. Invent no anatomy or accessories. Select signature inserts from its most recognizable details. Rigid parts rotate; flexible materials bend or stretch. A 15-second character-detail montage: exactly 20 shots joined by hard cuts. An empty, seamless warm-white studio background; soft directional light reveals the surfaces. Shots 1-19 are tightly cropped body details and rear angles: keep the face, eyes and facial profile entirely outside the frame, including reflections. Shot 20 is the first and only face reveal. No full-body view. Three connected bursts: a sharp head turn with the body bracing; a coiled arm opening into a strong sweep and torso twist; one short step into a planted pivot, finishing with a sudden look into camera. Carry body state across cuts without resetting to neutral. Existing hair, clothing and attachments react naturally. Keep the character active, with clear articulation and weight shifts. Jump between distant body regions. Mix brief visual punches with longer moving-detail reads. Use one abrupt camera move per shot except locked-camera inserts 4 and 11, where the character supplies the speed. Vary height, angle and direction. Brake moves briefly for legibility while the subject keeps moving. Local whips stay tightly cropped; changes of body region use hard cuts. Connect selected cuts through similar shapes or motion vectors, without morphing. Real-time action, no slow motion or freeze frames. 0-3s, shots 1-5. A concealed head turn triggers the first burst: 1. Extreme upper-back detail: a crash push catches its surface shifting as a sharp head turn begins entirely above the crop. 2. Fingers: a short downward snap catches them curling into a fist as the body braces. 3. Rear three-quarter shoulder detail: a tight camera arc catches the shoulder drawing back behind the turning head. 4. Locked floor-level foot detail: the grounded support foot braces, its raised heel pressing down firmly without taking a step. 5. Rear neck detail: a short upward tilt catches the nape stretching with the turn; stop below the back of the head, keeping the jaw and face excluded. 3-9s, shots 6-13. The coiled arm opens into a forceful sweep: 6. A signature upper-body detail: a fast diagonal slide catches an existing seam, joint or distinctive contour responding as the arm draws inward. 7. Elbow: a sharp lateral track catches the folding joint tightening the arm's coil. 8. Supporting knee: hard match cut from the bent elbow to the knee's similar angle; a short camera drop catches the leg compressing under the same preparation. 9. Wrist: a quick shallow orbit catches it turning outward to initiate the release. 10. Upper chest: a short punch-in catches the chest opening as the arm drives outward beyond the crop. Let this moving surface read slightly longer. 11. Locked hand detail: fingers spring open and the hand sweeps rapidly across the fixed frame. Catch one crisp, readable phase of the spread fingers before they leave the crop. 12. Upper back: a sharp sideways slide continues that screen direction as the shoulder rolls through the sweep and the torso twists. 13. A distinctive torso or waist detail: a close diagonal move catches its physical response to the twist, such as an existing fold pulling taut or an attached piece swinging. Give the material response a clear moving read. 9-13s, shots 14-19. One short step and a pivot build toward the reveal: 14. Thigh of the other leg: an abrupt low tracking move catches it driving the single short step out of the twist. 15. Advancing foot: a brief downward whip catches it landing; show clear floor contact. 16. Ankle and lower-leg detail: a fast side slide catches the landed leg accepting weight as the ankle flexes. 17. Waist: a tight orbit catches the torso rotating above the planted-foot pivot. The foot supports the turn rather than sliding across the floor. 18. Back-of-head contour: a short lateral whip catches the start of the return turn from behind. Keep only the rear contour in frame, never the facial profile. 19. A signature shoulder detail: a sharp diagonal slide catches its follow-through beneath the turning head, which stays completely above the crop. Cut at the peak of this movement. 13-15s, shot 20. First face reveal: 20. Hard cut to a frontal face close-up as the character sharply completes the turn toward the lens. An explosive short push reveals the entire face clearly for the first time and stops tightening before cropping out its features. Direct eye contact lands while hair, clothing or natural body follow-through remains alive. Keep the whole face readable through the end; finish on living motion, without a held pose or another cut. No added props, scenery, text, graphic overlays or split screens.

MiniMax H3X post · plays on click
Video0:15Watch the original
MiniMax H3Prompt

Reaper: a game equipment interface prototype

@ShamsAmin56 · 113 likes

Prompt - Claude Interactive Game Equipment UI (15s) Use Image 1 as a STRICT CHARACTER REFERENCE. Preserve the character exactly: oversized yellow hooded jacket, glossy black faceless helmet, black armored tactical torso, tattooed left arm, glossy red tactical cargo pants with hanging straps and pouches, black combat boots, and the massive crimson-black biomechanical right arm with organic tendrils and clawed fingers. Do not alter colors, proportions, outfit, or mechanical design. 0–2 Seconds High-angle overhead cinematic shot. The character sits on a vivid, highly saturated purple floor with one knee raised and the enormous crimson biomechanical arm resting beside him. He slowly tilts his head upward toward the camera. A premium AAA futuristic game interface fades in. Top left: Player Profile REAPER Level 42 Right side menu: START NEW GAME CONTINUE (highlighted) LOADOUT SETTINGS EXIT GAME A sleek white cursor smoothly clicks CONTINUE. 2–4 Seconds Camera pushes in toward the biomechanical right arm. A translucent RIGHT ARM EQUIPMENT panel slides in. Equipment list: PHANTOM GRIP ABYSS CLAW HELLFORGE TALON CRIMSON REAVER The selector moves to ABYSS CLAW. The right arm begins transforming. Red-black armor plates separate. Organic crimson tendrils weave together. Mechanical pistons rotate. Claw fingers extend into longer razor blades. Internal red energy pulses through exposed cables. Small particles and sparks emit as the transformation locks into place. 4–7 Seconds Camera smoothly arcs around the character's right side. A large ARMAMENT CUSTOMIZATION interface appears. Component grid: Shoulder Upper Arm Elbow Forearm Wrist Claw System Core Reactor Tendril Matrix Selections cycle rapidly. The biomechanical arm partially disassembles. Armor segments detach. Internal skeletal frame becomes visible. Hydraulic pistons rotate. Mechanical tendrils retract and regrow. New crimson armor slides into place. Claw geometry reshapes into a more aggressive configuration before locking with a satisfying mechanical snap. 7–8.5 Seconds Camera pulls back to a medium shot. A large glowing CONFIRM CONFIGURATION button appears. Cursor clicks. Every interface panel folds inward using holographic animations and disappears. The character rises from the seated position. He slowly flexes the upgraded biomechanical arm. The massive claws open and close once. Crimson energy briefly surges through the entire limb. 8.5–10 Seconds A cinematic LOADING bar appears across the bottom. Progress accelerates from 0% → 100%. The purple environment gradually fades into darkness. Golden light leaks through the scene. Subtle volumetric fog forms. Ambient audio transitions from UI sounds to a living city. 10–15 Seconds The world fully loads. The character now stands inside a dense cyberpunk megacity. Rain-soaked streets. Neon holographic advertisements. Steam rising from vents. Moving crowds. Hover drones patrol overhead. Motorcycles race past. Towering futuristic skyscrapers loom in the distance. Camera settles into a polished third-person gameplay perspective behind the character. HUD fades in: Minimap (top right) Health & Energy (bottom left) Ability Icons Weapon Indicator Objective Marker The character activates the crimson biomechanical arm, flexes the claws once, then confidently walks into the neon-lit street as the mission begins. Style: AAA Unreal Engine 5 cinematic, photorealistic PBR materials, production-quality game UI, premium sci-fi HUD, seamless mechanical transformation, cinematic camera motion, ray-traced lighting, volumetric fog, rain reflections, 4K, 60fps, no glitches, no UI clipping, no extra limbs, perfect character consistency, movie-grade color grading.

MiniMax H3X post · plays on click
Video0:15Watch the original
MiniMax H3Prompt

A reference-driven retro urban anime opening

@PixVerse · 73 likes

Here’s the prompt we used: subject_definitions: Character-feature reference: <image 1>:Character appearance reference <image2>: Environmental visual-style reference. Generate a 15-second, 16:9 landscape Japanese anime opening sequence. Use the visual language of a modern Japanese anime opening with a retro printed urban atmosphere. Combine 2D hand-drawn animation, cel shading, dynamic key animation, and graphic-design aesthetics. Preserve a tactile paper-grain surface, halftone dots, and retro manga print-registration misalignment caused by layered color printing. This should not be a character showcase PV. It should be a strongly rhythmic character montage. Center the sequence around one cool, retro urban girl, maintaining the exact same face throughout the entire video: jet-black bobbed hair with straight bangs and sharply layered ends, cool dark almond-shaped eyes, matte red lips, a signature beauty mark beneath the lower lip, a black tank top with heart-shaped patches, layered beaded necklaces, and a padlock pendant. Keep her facial structure and overall presence consistent in every shot. Use a three-color visual system dominated by black, lake blue, and scarlet red, with milk-white accents. Build her personality through large areas of solid color, abstract urban silhouettes, geometric color blocks, shadows, manga-style compositions, and graphic visual symbols. Use the rapid editing language commonly found in Japanese anime openings, including extreme close-ups of the eyes, red lips, padlock pendant, fingertips, and hair ends; tilted compositions; low-angle perspectives; sudden push-ins; frame slicing; and dynamic visual recombination. The editing rhythm may be freely arranged by the model according to the electronic beat. Add halftone patterns, ink lines, drifting heart-shaped stickers, halos and light spots, film grain, color-registration offsets, floating flat graphic text, and geometric symbols as part of the visual language. Emphasize a cool, distant, retro-urban loneliness with a mysterious, alluring quality, combined with a touch of City-pop nostalgia and a sense of fate. The result should feel like the official opening sequence of a Japanese television anime, not a game promotional CG trailer. Avoid complex fantasy environments, realistic locations, battle demonstrations, 3D rendering, and any drift in the character’s facial features.

MiniMax H3X post · plays on click
Video0:15Watch the original
MiniMax H3Prompt

VOID: a kinetic anime introduction

@TechieBySA · 37 likes

“Pure kinetic motion graphics from frame one. No scenes, no dialogue — just ink, energy and typography exploding across a void-black canvas. JJK-style aesthetic: ink splatter reveals, chromatic aberration pulses, impact freezes, speed lines. Music: Constant tension electronic — 160BPM mechanical hi-hats, layered detuned synths, staccato string stabs. No drop, no resolution. Just relentless building pressure the entire 15 seconds. 0:00–0:01: Black ink detonations consume a white frame. Negative space forms a fragmented "V." 0:01–0:03: Purple energy slash tears the frame open. White particles rush in and assemble into character silhouette KAI — his name explodes in massive white text with cyan underline, then shatters. 0:03–0:05: Cyan geometric shards flash rapidly, ghost-forming character SEREN. Her name slams in from off-screen, cracks the frame on impact, white light bleeds through. 0:05–0:08: Text fragments of V-O-I-D strobe across white void at different sizes and rotations. Character REN appears as a single ink brushstroke — impossibly calm. His name visible only as negative space within the ink. 0:08–0:10: Full graphic meltdown — rapid-fire collage of ink explosions, overlapping character names, speed lines, a single 6-frame red flash, each letter of VOID filling the entire frame one at a time. 0:10–0:13: Purple and cyan energy streams collide center-frame, compressed into the title VOID — ultra-bold, letters split purple-left cyan-right with white fracture lines running through each. Cast names slide in below. Pulses like it's alive. 0:13–0:15: Fracture lines spread, black consumes the letters from inside. Gone. Colors: #03020a black · #7c00ff purple · #00f0ff cyan · #ff1a1a red flash only · #ffffff fracture/text Never: soft transitions, static frames, gradients, faces, anything gentle.”

MiniMax H3X post · plays on click
Video0:24Watch the original
MiniMax H3Prompt

An elven rider: Seedance 2.5 and MiniMax H3 compared

@Tomw852 · 9 likes

subject_definitions: - @Image1: The young elven maiden warrior (S1) with sharp pointed ears, long platinum-blonde hair tied with fine forehead braids, wearing the navy blue long robe with ornate gold-embroidered trim. - @Image2: The desolate Nordic wilderness environment (S2) beneath gloomy overcast storm clouds with pale, cold ambient twilight on the horizon. summary: An intense, high-speed backward tracking long shot capturing the elven maiden (S1) riding a white warhorse at full gallop across the frozen tundra (S2), framed with naturalistic filmic realism, extreme speed streaks, and raw diegetic soundscape. retention_analysis: - @Image1: Retain exact elven facial bone structure, platinum blonde hair, forehead braids, and navy blue gold-trimmed robe. Adapt pose to active riding stance. - @Image2: Retain gloomy overcast sky, cold desaturated color palette, and barren tundra terrain while introducing dynamic motion blur. detailed_description: From 00:00.000 to 00:08.000, shot on an 85mm T1.8 cinema prime lens at eye-level framing a medium shot of the elven maiden (S1) astride a powerful white warhorse charging forward at extreme speed. The camera maintains a precise high-speed backward tracking movement 3.5 meters directly in front of the horse, locked in perfect speed synchronization while incorporating organic micro-breathing vibrations and subtle vertical road jostles authentic to naturalistic handheld cinematography. The lighting strictly mirrors @Image2: heavy, gloomy overcast clouds cover the sky, casting a soft, cold ambient atmospheric glow that gently rims S1's blonde locks and pointed ears, with shadows subtly lifted with a desaturated cyan-teal tint and authentic 35mm film grain. Dynamic focus is razor-sharp on S1's face and the horse's head: S1's cold, merciless amber eyes remain dead-locked onto the camera lens without blinking or smiling, her jaw set firm against the howling gale. Wild wind furiously whips her blonde bangs, fine braided hair, and the horse's white mane straight backward. The surrounding frozen ground and distant dark ridgelines stretch into radical, flat horizontal motion blur streaks, conveying blistering velocity. White vapor clouds burst rhythmically from the warhorse's nostrils into the frigid air. overall_soundscape: Zero non-diegetic background music (NO BGM). Pure, hyper-realistic diegetic soundscape: heavy, relentless galloping hoofbeats slamming into frozen gravel like war drums, brutal gale-force wind whistling and buffeting the camera microphone, accompanied by deep, rapid, guttural horse exhalations and the subtle jingling of leather bridle hardware. non_diegetic_music: None. Complete silence of musical score to maximize raw, cinematic documentary tension.

02How-to

How to Write MiniMax H3 Prompts

These steps follow the MiniMax H3 documentation and official prompt guide, linked in the sources section. H3 supports text-to-video, image-to-video with a first and/or last frame, and reference generation using images, video and audio.

01

Choose the mode and output

Choose text-to-video, first- and/or last-frame image-to-video, or reference generation. H3 outputs 768P or 2K video for an integer duration from 4 to 15 seconds, with a prompt limit of 7,000 characters.

02

Separate the scene, soundscape and music

The base prompt guide organizes the prompt into integrated_multimodal_description, overall_soundscape and non_diegetic_music. Use those sections to describe the audiovisual scene, its overall sound and its music.

03

Define what each reference guides

Reference mode includes subject_definitions, summary and detailed_description sections. References can guide characters, motion, camera, style, voice and editing rhythm. The limits are 9 images, 3 videos and 3 audio clips, with at most 12 mixed files.

03Copyable · quoted verbatim

MiniMax H3 Prompts from Public Sources

Template · A reference-led character detail montageQuoted verbatim

Use @[character reference] for the character's identity, anatomy, costume, colors, materials and visual style. Adapt body-region cues to its actual anatomy, replacing absent features with existing equivalents. Invent no anatomy or accessories. Select signature inserts from its most recognizable details. Rigid parts rotate; flexible materials bend or stretch.

A 15-second character-detail montage: exactly 20 shots joined by hard cuts. An empty, seamless warm-white studio background; soft directional light reveals the surfaces. Shots 1-19 are tightly cropped body details and rear angles: keep the face, eyes and facial profile entirely outside the frame, including reflections. Shot 20 is the first and only face reveal. No full-body view.

Three connected bursts: a sharp head turn with the body bracing; a coiled arm opening into a strong sweep and torso twist; one short step into a planted pivot, finishing with a sudden look into camera. Carry body state across cuts without resetting to neutral. Existing hair, clothing and attachments react naturally. Keep the character active, with clear articulation and weight shifts.

Jump between distant body regions. Mix brief visual punches with longer moving-detail reads. Use one abrupt camera move per shot except locked-camera inserts 4 and 11, where the character supplies the speed. Vary height, angle and direction. Brake moves briefly for legibility while the subject keeps moving. Local whips stay tightly cropped; changes of body region use hard cuts. Connect selected cuts through similar shapes or motion vectors, without morphing. Real-time action, no slow motion or freeze frames.

0-3s, shots 1-5. A concealed head turn triggers the first burst:
1. Extreme upper-back detail: a crash push catches its surface shifting as a sharp head turn begins entirely above the crop.
2. Fingers: a short downward snap catches them curling into a fist as the body braces.
3. Rear three-quarter shoulder detail: a tight camera arc catches the shoulder drawing back behind the turning head.
4. Locked floor-level foot detail: the grounded support foot braces, its raised heel pressing down firmly without taking a step.
5. Rear neck detail: a short upward tilt catches the nape stretching with the turn; stop below the back of the head, keeping the jaw and face excluded.

3-9s, shots 6-13. The coiled arm opens into a forceful sweep:
6. A signature upper-body detail: a fast diagonal slide catches an existing seam, joint or distinctive contour responding as the arm draws inward.
7. Elbow: a sharp lateral track catches the folding joint tightening the arm's coil.
8. Supporting knee: hard match cut from the bent elbow to the knee's similar angle; a short camera drop catches the leg compressing under the same preparation.
9. Wrist: a quick shallow orbit catches it turning outward to initiate the release.
10. Upper chest: a short punch-in catches the chest opening as the arm drives outward beyond the crop. Let this moving surface read slightly longer.
11. Locked hand detail: fingers spring open and the hand sweeps rapidly across the fixed frame. Catch one crisp, readable phase of the spread fingers before they leave the crop.
12. Upper back: a sharp sideways slide continues that screen direction as the shoulder rolls through the sweep and the torso twists.
13. A distinctive torso or waist detail: a close diagonal move catches its physical response to the twist, such as an existing fold pulling taut or an attached piece swinging. Give the material response a clear moving read.

9-13s, shots 14-19. One short step and a pivot build toward the reveal:
14. Thigh of the other leg: an abrupt low tracking move catches it driving the single short step out of the twist.
15. Advancing foot: a brief downward whip catches it landing; show clear floor contact.
16. Ankle and lower-leg detail: a fast side slide catches the landed leg accepting weight as the ankle flexes.
17. Waist: a tight orbit catches the torso rotating above the planted-foot pivot. The foot supports the turn rather than sliding across the floor.
18. Back-of-head contour: a short lateral whip catches the start of the return turn from behind. Keep only the rear contour in frame, never the facial profile.
19. A signature shoulder detail: a sharp diagonal slide catches its follow-through beneath the turning head, which stays completely above the crop. Cut at the peak of this movement.

13-15s, shot 20. First face reveal:
20. Hard cut to a frontal face close-up as the character sharply completes the turn toward the lens. An explosive short push reveals the entire face clearly for the first time and stops tightening before cropping out its features. Direct eye contact lands while hair, clothing or natural body follow-through remains alive. Keep the whole face readable through the end; finish on living motion, without a held pose or another cut.

No added props, scenery, text, graphic overlays or split screens.
MiniMax H3@aimikoda799 wordsOpen case →
Template · An 8-bit black-and-white San Francisco animationQuoted verbatim
8-bit, black and white animation of san francisco
MiniMax H3@vercel_dev8 wordsOpen case →

04Sources

Official Sources

05Related topics

Related topics

06FAQ

Frequently Asked Questions

What can MiniMax H3 generate?

MiniMax describes H3 as an open, general-purpose multimodal video model that understands text, image, video and audio. It supports video generation, reference-based creation and video editing. MiniMax H3 Max is a faster variant post-trained by fal.ai, with 480P and 768P output.

What are the output and input limits?

Output is 768P or 2K and lasts an integer number of seconds from 4 to 15. A prompt can contain up to 7,000 characters. Reference input supports up to 9 images, 3 videos and 3 audio clips, with at most 12 files in a mixed set.

How does the official prompt guide organize a prompt?

The base guide uses integrated_multimodal_description, overall_soundscape and non_diegetic_music. Reference mode adds subject_definitions, summary and detailed_description sections. References can guide characters, motion, camera, style, voice and editing rhythm.