APIMODELSAPIMODELS

MiniMax H3 Video Editing Prompts

Curated, copy-ready MiniMax H3 (Hailuo 3) video prompts, each with a sample clip and credit to its author. Hover to play, one click to copy.

← All 226 MiniMax H3 prompts

Last updated: · MiniMax H3 API — pricing & playground · API docs

TL;DR

MiniMax H3 (community name: Hailuo 3.0) is the multimodal video model MiniMax released on 31 July 2026: 2K, 24fps, 4-15 second clips with natively synced audio and dialogue, supporting text-to-video, first/last-frame control, reference images and videos, motion transfer and generative video editing — up to 9 reference images, 3 video clips and 3 audio clips per request. This page collects copy-ready H3 prompts, each with a sample clip and credit to its author, free to use.

Source: MiniMax (the model vendor’s official site)

Hover to preview · click for sound

3x3 contact sheet female rapper grid performance

Use @Image 1 as the exact opening frame. It is a 3x3 contact sheet of nine separate portraits of the same female rapper, each in a different location. Treat every panel as its own live video cell with its own camera, lighting and performance. 12 seconds, 16:9. Keep the grid absolutely fixed for the entire shot: identical panel positions, identical panel sizes, identical white gutters, no zoom, no pan, no rotation, no drift of the grid itself. Only the content inside each panel may move. Preserve each panel's identity exactly: same face, same hair, same wardrobe, same jewellery, same background as in @Image 1. Panel 1 stage with mic, panel 2 graffiti alley, panel 3 studio console with bandana, panel 4 rooftop with pink car, panel 5 roller shutter with bucket hat and chains, panel 6 purple club, panel 7 sunset rooftop with braids, panel 8 neon alley, panel 9 black backdrop close-up. [0.0-1.2s] All nine panels sit almost still like breathing photographs, only micro-movement: hair drift, blinking, smoke, light flicker. A low sub-bass swell rises. [1.2-3.0s] The beat drops. Panel 1 raps to camera: "MiniMax." Panel 5 answers: "H3." The other seven nod on the beat. [3.0-5.0s] Panels 2, 3 and 7 trade the line one after another: "New drop." Each one takes the beat, the rest bounce in rhythm. [5.0-7.5s] Panels 4, 6, 8 and 9 answer: "Go and try it." Hands move, chains swing, ad-libs. [7.5-10.0s] All nine perform in unison, tightly lip-synced toH3, go and try it." [10.0-11.3s] The beat stops dead. In one synchronized move all nine raise a hand into a finger-gun aimed straight at the camera, chin down, hard contact. [11.3-12.0s] They fire the imaginary shot on a single low hit, a tiny recoil ripples through all nine panels at once, then freeze. Audio: one continuous hip-hop beat across all panels, roughly 92 BPM, deep 808 sub, crisp trap hats, sparse hand claps. Female voice, confident, slightly playful, not aggressive. Layer nine slightly detunedn section for a choir effect. End on one low hit plus a shorttail of room tone. Keep the lip-sync accurate to the words in every speaking panel. No subtitles, no on-screen text, no captions, no logos, no watermarks. No panel may drift, crossfade, resize or swap places. Do not add people. Dreen single shot at any point.

Hover to preview · click for sound

Hard sci-fi desert 3x3 grid animation sequence

Epic desert sci-fi, harsh golden-hour light, volumetric sand haze, anamorphic lens flares, subtle handheld drift, heavy atmospheric particles, 35mm film texture. Use the provided 3x3 grid as the only visual reference. Lock the exact same woman (face, scars, green eyes, headscarf, desert armor, cape) and environments from every panel. Do not invent new outfits or faces. [TIMELINE] 0-1.7s: [Wide establishing] Lone figure walking across endless dunes toward twin massive suns. Heat shimmer, long shadow stretching behind her. 1.7-3.4s: [Extreme close-up] Her scarred face fills the frame. Eyes locked on camera, wind tugging the dirty headscarf, grit on skin, unblinking intensity. 3.4-5.1s: [Tight detail insert] Gloved hand grips the rusted multi-tool weapon. Metal catches the orange light, cable dangling, fingers adjusting the grip with quiet precision. 5.1-6.8s: [Low wide silhouette] The towering industrial structure rises against the giant sun. Dust storms swirl at its base, scale feeling ancient and oppressive. 6.8-8.5s: [Medium tracking] She advances through thick sandstorm, lower face wrapped, cape and scarf whipping hard. Eyes visible above the cloth, determined stride. 8.5-10.2s: [Hero wide] She stands alone on the rocky outcrop, cape flowing, surveying the empty desert below. Wind howls, silhouette sharp against the hazy sky. 10.2-11.9s: [Dynamic tracking + low angle] She charges down the steep dune face, boots kicking up thick clouds of sand, body leaning forward with full momentum. 11.9-13.6s: [Macro atmospheric] Golden sand particles explode and float in extreme slow motion, backlit by the blazing sun, pure texture and light. 13.6-15s: [Intimate close-up] Inside the dark cave, shafts of light cut through. She sits against the rock, armor dusted, breathing controlled, eyes sharp and watching the entrance. Photorealistic, ultra-detailed skin and fabric textures, perfect sand physics, coherent character identity across every cut, cinematic color grade, heavy film grain, no artifacts, movie-level continuity.

Hover to preview · click for sound

Boundless Journey Across Surreal Worlds

Reconstructed

Duration: 15 seconds | Aspect ratio: 16:9 | Style: Cinematic surreal environment transition montage SCENE Young woman, dark wavy hair, white satin dress with a high slit, camel blazer draped over shoulders, emerald and diamond necklace. She travels seamlessly through rapidly changing fantastical environments. SHOT BREAKDOWN 0-4s — Woman walks confidently through a rainy cyberpunk city with glowing neon billboards and flying vehicles, transitioning instantly to a traditional Japanese temple with falling pink cherry blossoms. 4-8s — The environment shifts to an underwater realm featuring large sharks, glowing jellyfish, and massive crystal formations. The scene cuts to a gray lunar base with astronauts; the woman turns away, facing a giant Earth in the sky. 8-12s — Seen from a profile angle, she stands in a foggy mystical forest surrounded by giant bioluminescent mushrooms. The backdrop abruptly changes to a dark volcanic landscape with actively erupting lava and smoke. 12-15s — She faces forward inside a modern minimalist penthouse with floor-to-ceiling windows overlooking a sea of clouds at sunset. The final cut places her in deep space against a brightly lit spiral galaxy and stars. CAMERA Locked-on tracking shot, maintaining medium-full framing. Rapid, seamless smash cuts between backgrounds while the subject's position remains relatively stationary. LIGHTING & PALETTE Highly dynamic and contrasting. Neon magentas and cyans, warm golden hour tones, deep aquatic blues, fiery orange lava, and vivid cosmic purples. AUDIO Driving electronic beat. Heavy swooshing transition sound effects synchronized with the rapid environmental cuts. No dialogue. AVOID Wardrobe changes, character morphing, camera shake, shallow depth of field obscuring backgrounds, muted colors.

video by @ZephyraLeighSource
Hover to preview · click for sound

Cyber Grunge K-Pop MV

15s K-pop 女团 MV 提示词 主体:三人 K-pop 女团,人物形象与当前角色设定一致。 SOL:黑色长直发,冷静强势,黑色结构短西装、低腰百褶短裙、黑色长靴。 LUNA:银灰色短狼尾,冷感疏离,银灰短款机能夹克、黑色连体内搭、不对称裙裤、厚底靴。 CORONA:深红棕长卷发,带细辫,叛逆锋利,红黑赛车短夹克、低腰裙裤、绑带长靴。 场景:白色无缝影棚被重新处理成地下音乐杂志拍摄现场。背景不是干净商业棚拍,而是带有过曝白墙、灰黑阴影、复印纸贴片、撕裂纸边、半色调网点、胶带痕迹、扫描错位和局部黑色喷漆纹理。部分镜头中背景瞬间切换为纯黑、过曝白或高反差灰色,只通过硬切完成。 视觉风格:dark-pop、cyber-grunge、rap 音乐视频,写实高时装质感,90 年代末至 00 年代初独立杂志、地下音乐海报、复印纸、胶片扫描与 zine 拼贴美学。高反差黑白与低饱和暗红、银灰点缀,肤色真实,黑位厚重但保留衣服细节。粗颗粒、轻微胶片抖动、曝光跳动、半色调网点、印刷毛边、扫描偏移、局部色版套印错误。不要廉价赛博霓虹,不要游戏 CG,不要过度磨皮。 摄影:真实音乐视频拍摄感,24mm、35mm 和 50mm 镜头快速切换。近距离广角压迫感、头部点拍、肩部特写、低机位半身、手部推进镜头、贴脸侧面特写。镜头可以轻微手持、突然向前冲、快速横移或在鼓点上瞬间停住。人物表演卡住鼓点,不是走秀,不是静态摆拍。 音乐:15 秒 dark-pop / cyber-grunge / trap rap 段落。低频 bass 强烈,干脆 snare,失真电子噪音,短促人声切片。节奏约 138 BPM。前 3 秒建立压迫感,中间进入连续 rap,最后 2 秒重低音收束。 文字包装:文字不是普通字幕,而是画面主体的一部分。使用原创无品牌词组。字体混合粗黑无衬线、窄体工业字、打字机字和手工复印字。文字位于人物前景、中景与背景之间,人物可以遮挡部分文字,文字也可以压过身体、服装或手部,但不能遮挡眼睛和主要面部表情。文字跟随 rap 重音、snare 和 bass hit 瞬间出现、震动、拉伸、压缩、断裂、翻转或硬切重组。文字边缘带油墨溢出、复印噪点、色版错位和扫描毛边。 剪辑规则:全片只使用硬切、跳切、闪白切、闪黑切和鼓点匹配切。禁止淡入淡出,禁止溶解,禁止柔和转场,禁止慢动作堆叠。每个镜头 0.3 至 1.2 秒,快速但清晰,动作和文字都精准卡住节拍。 SHOT 01|0.0–0.8s 纯黑画面在 bass hit 上硬切为 SOL 的极近距离正面特写。24mm 广角贴近面部,身体前倾靠近镜头,头部突然向下点拍,黑色长发轻微甩动。画面高反差,背景过曝白。巨大粗黑文字位于她身后,头部遮挡文字中间部分。胶片颗粒突然增强,画面有一次轻微上下抖动。 SHOT 02|0.8–1.5s 硬切 LUNA 肩部与下颌特写。她压低肩膀,侧头看向镜头,在 snare 上突然把脸转正。银灰短发扫过眼角,但不遮挡眼睛。半色调网点从背景切入人物衣服边缘。窄体文字从画面右侧被硬切压缩进来,位于她身后与肩膀前景之间。 SHOT 03|1.5–2.3s 硬切 CORONA 低机位半身。她一只手快速推向镜头,手掌掠过镜头前景,在 bass hit 上收回,身体向前压。深红棕头发带有真实惯性。红黑色版轻微错位,画面像旧杂志扫描。文字被她的手部遮挡一部分。 SHOT 04|2.3–3.0s 三人正面中景,站位紧凑。SOL 位于中央,LUNA 和 CORONA 分居两侧。三人同时肩膀下压,再在鼓点上抬头看镜头。背景瞬间从白色硬切为黑色,再切回白色。大型文字在人物背后横向拉伸,人物身体遮挡文字中部。 SHOT 05|3.0–3.7s SOL 50mm 侧面近景。她转身时突然停住,脸从侧面转向镜头,手指沿着下颌快速划过。镜头在动作停点时突然向前推进半步。文字位于前景左侧,轻微压过她的肩膀,但不遮挡脸。 SHOT 06|3.7–4.4s LUNA 全身低机位。她做一个短促、真实的舞蹈动作:膝盖内扣、上半身向后拉、手臂在胸前切开,然后立即向镜头前倾。银灰外套与透明袖套产生清晰动态。画面在 snare 上切成黑白半色调 3 帧,再恢复彩色。 SHOT 07|4.4–5.1s CORONA 35mm 贴脸特写。她侧头、咬住节奏轻微抬下巴,随后向镜头靠近。不是微笑,而是带挑衅感的真实 rap 表情。暗红文字从她身后突然切到前景,压住头发和夹克局部,不遮挡眼睛。 SHOT 08|5.1–6.0s 三人快速交替头部点拍 montage。SOL 正面点头、LUNA 侧面甩头、CORONA 向前压头。每个动作只保留 3 至 5 帧,用三个连续硬切完成。背景在白、灰、黑之间切换。胶片扫描线短暂上下滚动。 SHOT 09|6.0–6.8s SOL 和 CORONA 同框近景。SOL 正面对镜头 rap,CORONA 从她身后突然探出并靠近镜头。两人的前后层次明确。文字分别位于背景与前景下方,压过服装但不遮脸。 SHOT 10|6.8–7.6s LUNA 单人中近景。她双手在脸侧快速交叉,然后向两边切开,动作卡住连续 snare。镜头同步快速向左横移,再在最后一个重音瞬间停住。背景出现撕裂复印纸和黑色胶带图形,像真实贴在棚墙上。 SHOT 11|7.6–8.5s 三人全身广角。她们不是整齐跳舞,而是分别完成不同但互相呼应的动作:SOL 向前迈一步并肩膀下压;LUNA 身体向侧面折叠;CORONA 转身后回头靠近镜头。动作在同一个 bass hit 上同时停住。 SHOT 12|8.5–9.3s 极近距离手部 montage。金属戒指、腰链、夹克拉链、手指敲击腿部、手掌推向镜头。每个细节只出现 2 至 4 帧。文字碎片像剪贴纸一样在不同画面位置硬切重组。 SHOT 13|9.3–10.2s CORONA 在前景 rap,镜头使用 24mm 靠得很近。她一只手指向镜头,然后手腕旋转向下压。LUNA 和 SOL 在后方虚化,但仍做小幅头部与肩部动作。人物与文字形成三层关系:背景文字、中景两名成员、前景 CORONA 和手势。 SHOT 14|10.2–11.1s LUNA 从画面左侧快速转入镜头。她转身时衣摆甩动,随后脸贴近镜头侧边,眼睛直视镜头。画面在重音上出现轻微水平扫描错位,人物轮廓产生红青色版偏移,但肤色保持真实。 SHOT 15|11.1–12.0s SOL 中央正面半身。她低头停顿半拍,随后猛然抬头,三人从她身后同时向前压近。镜头也向后快速退一步,制造真实空间冲击。巨大文字从背景突然切到人物前景之间,被三人的身体切割成多个部分。 SHOT 16|12.0–13.0s 连续三次硬切:SOL 手指推镜头;LUNA 肩膀撞向镜头方向;CORONA 转身回头。每个镜头都配合一个 snare。画面颗粒和印刷噪点逐次加重。 SHOT 17|13.0–14.0s 三人低机位全身英雄镜头。SOL 中央向前站一步,LUNA 侧身压低重心,CORONA 身体前倾。三人保持真实呼吸和微小晃动,不是冻结静态 pose。背景大字被人物遮挡,只能看到分裂的字形。 SHOT 18|14.0–15.0s 最后一个 bass hit。镜头突然推进到三人的脸部近景,三人同时头部点拍并直视镜头。画面瞬间压缩成高反差黑白复印效果。原创文字短暂出现在人物背后与肩膀前景之间,轻微震动两次。最后直接硬切纯黑,不淡出。 负面要求:不要普通女团排排站,不要持续走秀,不要静态时尚 pose,不要软弱缓慢动作,不要过度机械舞蹈,不要卡通表情,不要赛博朋克霓虹城市,不要廉价 RGB 光污染,不要真实品牌或任何现实中存在的组合名与歌曲名,不要普通字幕,不要淡入淡出,不要溶解转场,不要柔光梦幻滤镜,不要游戏 CG 感,不要过度平滑的 AI 运镜。

by @liandeli2Source
Hover to preview · click for sound

DV Camcorder Gym Vlog POV

CAMERA: DV 16mm tape camcorder POV. CHASE films herself, occasionally propping the camera on a wall or gym bag. Natural handheld shake, imperfect framing, delayed focus, clumsy zooms, face occasionally cut off. Camcorder never visible. LOOK: Soft blurry tape aesthetic, faint tape noise, bloomed gym lights, flickering auto-exposure, muted contrast, realistic skin. STYLE: Calm post-workout gym vlog. Relaxed, reflective, slightly tired but happy. Slow pacing, natural laughs. CHARACTER: CHASE, Korean idol (20s). Long black hair loosely tied, light post-workout glow, slim athletic build. Modest long-sleeve athletic top, loose joggers/leggings, socks or shoes off, towel around neck, no jewelry. SETTING: Quiet evening gym. Stretching mat, mirror wall, gym bag, protein shake, soft lighting, mostly empty. 15s | 6 Cuts Propped medium: sits on mat, exhales. "Okay, cool-down time." Handheld stretch: hamstring stretch, laughs. "Yeah, that was a lot today." Macro: picks up protein shake, condensation visible. Ambient sound only. Medium: drinks shake, smiles. "Okay, that's actually really good." Close: overhead stretch, rates workout. "I'd give today like an eight out of ten." Selfie: grabs bag and shake, waves. "Alright, I'm heading out—see you guys."

by @doctorwasifSource
Hover to preview · click for sound

Avant-Garde Kinetic Fashion Lookbook

Reconstructed

Duration: 15 seconds | Aspect ratio: 16:9 | Style: High-fashion editorial lookbook with bold kinetic typography SCENE Female model with sleek hair wearing avant-garde structural fashion, including oversized blazers, metallic armor, and sharp-shouldered dresses. Backgrounds alternate between concrete brutalism, glowing metallic tunnels, and minimalist photo studios. Giant text overlays integrate with the environment. SHOT BREAKDOWN 0-4s — Model in an oversized black suit poses against concrete pillars and a red industrial corridor. Framed center. Floating text reads: "DON'T BLINK" and "BREAK THE RULES". 4-8s — Model struts through a red metallic tunnel in a white and silver futuristic corset, then poses in a stark split-tone studio. Framed medium-wide. Text reads: "NO APOLOGIE" and "MAKE THEM". 8-12s — Model wears a reflective silver top against a dark glitch backdrop, then walks strictly in profile in a white and silver suit across a seamless white void. Text reads: "6411", "LOUDER LOUDER LOUDER", and a halftone "NO". 12-15s — Model poses in a sharp-shouldered emerald green outfit in front of a torn-poster wall, then stands motionless within a dismantled photo studio setup. Wide framing. Text reads: "WE DON'T MISS", "4:0:21.18.1 1:2:3:8", "OWN THE NIGHT", and "UNSTOPPABLE". CAMERA Locked-off center framing, strict linear tracking, rapid rhythmic smash cuts. LIGHTING & PALETTE High-contrast studio flash, harsh directional spotlights. Dominant palette: stark white, deep black, metallic silver, crimson red, emerald green. AUDIO Heavy electronic bass beat, rhythmic camera shutter clicks, sharp metallic swooshes. AVOID Soft focus, warm natural sunlight, organic environments, slow crossfades, shaky handheld camera.

video by @zahra4sureSource
Hover to preview · click for sound

Surreal Crimson Sports Fusion

Reconstructed

Duration: 15 seconds | Aspect ratio: 9:16 | Style: High-energy, surreal sports montage SCENE Multiple athletes including a female soccer player in a green tracksuit, a male golfer in a red jacket, and a shirtless male basketball player in red shorts, situated in abstract, dark, and dramatically lit environments. SHOT BREAKDOWN 0-4s — A woman with braids in a green and yellow jacket stares intensely, then walks forward in a dark stadium. A white soccer cleat kicks a yellow and red ball. 4-8s — A massive soccer ball dominates the foreground. Rapid cuts reveal a golfer mid-swing, extreme close-ups of a gloved hand gripping a club, and the textured dimples of a golf ball. 8-12s — A golfer wearing yellow sunglasses and a white cap reading "SUPER" swings, spraying debris. The golf ball flies into a pitch-black void. 12-15s — A shirtless basketball player raises a ball against a twilight sky, then leaps into the air with glowing red motion trails. The ball falls onto grass before the scene distorts into a melting, abstract red sphere. CAMERA Kinetic pacing with rapid cuts. Mix of extreme macro close-ups, low-angle hero shots, and dramatic slow-motion action freezes. LIGHTING & PALETTE High-contrast, directional lighting with intense crimson red gels and deep shadows. Primary palette features emerald green, vivid red, stark white, and deep teal. AUDIO Heavy synthetic bass, aggressive swooshes, sharp impact sounds of sporting equipment, and distorted low-frequency hums. AVOID Muted colors, flat lighting, continuous long takes, realistic physics, mundane settings.

video by @ibexdreamSource

How to write great MiniMax H3 prompts

  1. 1One action per prompt. H3 clips are 4-15 seconds. "She pushes the door open and glances back with a smile" beats "she opens the door, crosses the hall, sits down and starts typing" — the second gets compressed into mush.
  2. 2Spell out the camera move. Dolly in, pull back, pan, tracking shot, handheld drift, drone descent. Camera language is the biggest difference between video and image prompts; leave it out and you get a locked-off tripod.
  3. 3Write the sound — H3 generates it. H3 produces synced native audio. Name the ambience (rain, subway announcement), the action sound (heels on tile), even a line of dialogue, and the clip lands a whole tier more finished.
  4. 4Set light and palette first. "Cold blue night with neon spill" or "warm golden-hour backlight" — one lighting/color sentence decides the whole look, exactly as in image prompts.
  5. 5Give it speed and timing. Slow motion, real time, timelapse — and when the beat happens. Without a rhythm cue the model spreads the action evenly and it reads sluggish.
  6. 6References carry identity, the prompt carries action. H3 takes up to 9 reference images. Let them hold the face, wardrobe or product; keep the prompt focused on what happens and how the camera moves.
  7. 7State aspect ratio and the final beat. 16:9 for landscape, 9:16 for short-form. Adding where the shot ends ("hold on a close-up of the product") saves you an edit later.

Prompt formula

Subject + specific action + camera move + lighting & palette + environment detail + sound/dialogue + pacing + aspect ratio

Example: Outside an izakaya on a rainy night, a woman in a beige trench closes her umbrella and turns; the camera dollies slowly in from over her shoulder to a profile close-up. Cold blue streetlight mixed with warm orange lantern glow, rain beading on the umbrella. Rain ambience and a distant train rumble; she says quietly, "You are late." Real time, 16:9.

Frequently asked questions

What is MiniMax H3?

MiniMax H3 (commonly called Hailuo 3.0) is the multimodal video model MiniMax released on 31 July 2026. It outputs 2K video at 24fps, 4-15 seconds long, with natively generated synced audio and dialogue. It supports text-to-video, first/last-frame control, reference images and videos, motion transfer and generative video editing — up to 9 reference images, 3 video clips and 3 audio clips in one request.

Are these prompts free?

Yes. Every prompt here is free to copy and run anywhere you have access to H3 — no login, no payment.

Who wrote these prompts?

Each card credits the prompt author, and the credit links to the original post or profile. We only curate; copyright stays with the author. If you are an author and want a prompt removed or a credit corrected, email us.

Where can I run these prompts?

H3 is live on apimodels.app as the model minimax-h3 — native 2K with synchronized audio, any whole length from 5 to 15 seconds, and up to 9 reference images plus 3 reference videos and 3 audio clips in one request. It is billed per second in two resolution tiers — native 2K at $0.145/s (5 seconds is $0.725) and 768P at $0.097/s (5 seconds is $0.485) — the first 5 reference images are free and each extra image is $0.054, and only successful requests are charged. The same key and the same endpoint also cover Seedance, Kling, VEO and Grok. You can also run H3 on the official MiniMax API or in the Hailuo product.

How do I write a good H3 prompt?

Write each prompt like a single storyboard card: what the subject does, how the camera moves, the lighting and palette, the environment details — then the sound you want, since H3 generates native audio and dialogue. Keep it to what fits in 4-15 seconds; one clear action beats three crammed beats.

Can I use H3 videos commercially?

That is governed by MiniMax's terms of use — check the official terms. The prompt text collected here is yours to edit and remix freely.

Run these prompts on apimodels.app

MiniMax H3 is live here alongside 180+ other image, video, audio and language models — one API, one key, pay as you go, with $0.10 of trial credit on sign-up.

This library is open source on GitHub

The prompts we reconstructed are published in full under MIT; author-written prompts are indexed with credit and a link to the original post. Open an issue to suggest more.

View on GitHub

Other prompt libraries