← MiniMax H3 prompt guide

Animated styles and art direction

Pottery, paper cut-outs, woodblock prints, watercolour and pixel art, each with its own rules of motion.

The templateAll 23 clips

A watercolour calendar card in ten shots

Shot list · MiniMax H3 per the author · Sep 9, 2026

Ten shots of 1.5 seconds each hold one image and one caption apiece, so the shot list and the text ledger are the same list. The style line fixes medium, palette and handwritten type once at the top, and the soundscape names things that appear in the shots, such as baskets, footsteps and pouring juice. With no characters and no camera instructions, little has to stay consistent from shot to shot, which suits a recurring daily post.

Animated styles and art directionKinetic type and title sequences

Read @airina_xyz's prompt
Create a 15-second multi-shot short video in a cozy watercolor illustrated style with vineyard warmth, painterly fruit clusters, and old harvest tradition charm. The palette should include vineyard green, dusky purple, golden sunlight, cream, and earthy brown. Include on-screen text in a soft handwritten style. Soundscape: vineyard ambience, baskets, soft footsteps, juice pouring.

Shot list

0.0–1.5s — Close-up of grape clusters glowing in sunlight.
Text on screen: “September 9”

1.5–3.0s — Rows of vineyard vines stretching into the distance.
Text on screen: “Grape harvest”

3.0–4.5s — Hands snip grapes from the vine and lower them into baskets.
Text on screen: “Harvest time”

4.5–6.0s — Deep purple and pale green grapes piled together.
Text on screen: “Sweet clusters”

6.0–7.5s — Grape juice or must being poured into a rustic glass.
Text on screen: “Fresh must”

7.5–9.0s — A table with grapes, leaves, and a simple seasonal dessert.
Text on screen: “Desserts of the season”

9.0–10.5s — Vineyard leaves begin to turn yellow at the edges.
Text on screen: “Summer becoming autumn”

10.5–12.0s — Basket carried through the vines in warm late-afternoon light.
Text on screen: “Old harvest traditions”

12.0–13.5s — A rustic still-life with grapes and a cloth napkin.
Text on screen: “Abundance in every cluster”

13.5–15.0s — Wide vineyard sunset view.
Text on screen: “September in the vineyard”

Original post ↗

What holds it together

A strong style prompt does more than name a look. It opens with a style block (medium, line, palette, texture, lighting) and then states how things are allowed to move in that medium: stop-motion jumps and locked wide shots for paper cut-outs, flat profile combat for red-figure pottery, ink bleeds for watercolour, sprite swaps for pixel art. The most consistent author in this set uses the same frame every time: one line naming the style with duration and aspect, an audio line, then four short acts with second ranges, ending on a reveal that comments on the medium, such as the whole scene turning out to be painted on a vase. Authors who start from an illustration reference ask for the exact visual language of that image and often return to it as the final frame. What goes wrong: styles that slide toward photoreal or generic 3D, smooth camera moves in media that should jump, and too many events for 15 seconds.

Write these in

  1. Medium, line quality, palette and texture in the first sentence
  2. How motion works in this medium: jumps, bleeds, flat profiles, sprite swaps
  3. Camera rules that fit the medium: locked wides, slow drifts, or none
  4. Sound that belongs to the medium: paper, brush, music box, koto
  5. About four acts with second ranges, one event each
  6. What it must never become: photoreal, 3D, or another style

What breaks it

  • Naming the style once and then writing live-action camera moves. Match the camera to the medium.
  • Forgetting to rule out photoreal or 3D. Styles drift toward the model's default look.
  • Packing a whole epic into 15 seconds. Pick four beats.
  • Starting from an illustration without saying what to keep. List line, palette, faces and proportions.

The template

[MEDIUM] animation, [DURATION] seconds, [aspect]. [Line quality], [palette], [texture: paper grain / glaze crackle / dithering], [lighting]. Never photoreal, never [other style to avoid].
Motion rules: [hard stop-motion jumps / ink bleeds between scenes / flat side-on movement]. Camera: [locked wide shots / slow drift / no camera moves].
Audio: [instruments and medium sounds]. [No voice / one narrator line].
Act 1 (0-[X]s): establish the world and [the hero] in this medium.
Act 2 ([X]-[Y]s): [a transition native to the medium]. [The turn].
Act 3 ([Y]-[Z]s): [the main event].
Act 4 ([Z]-end): [a reveal that comments on the medium, e.g. the scene is painted on an object]. Final [sound].

Ours to use freely. Replace everything in brackets.

23 clips that use it

Each plays from the creator's own post on X. Our note sits under it; the prompt belongs to them.

A motel suitcase that packs itself in stop-motion

JSON structure · MiniMax H3 per the author · Sep 1, 2026

The JSON defines stop-motion as a physical rule (objects jump a few centimetres per frame and never slide or float) and locks the camera, so every change on screen comes from the objects. The packing order is fixed in three groups, small, middle and bulky, with one rhythm block each, and the suitcase has to look full before the lid closes in three hits. Readable text is limited to three props, and the audio lists per-frame ticks and thumps while ruling out music, voices and cartoon effects.

Animated styles and art directionCamera moves and transitionsKinetic type and title sequences

Read @techhalla's prompt
{
  "archetype": "Stop-motion / frame-by-frame jitter of real objects moved between frames",
  "concept": {
    "title": "ROOM 214",
    "world": "A roadside motel room in Flagstaff, Arizona, late afternoon. Brass ROOM 214 on the interior door. Desk notepad stamped DESERT INN / FLAGSTAFF. Navy hard-shell suitcase open on the bedspread. Sun stripes through vertical blinds. Vacation clothes already scattered: socks, underwear, t-shirt, oxford shirt, jeans, coiled belt, knit sweater, windbreaker, cap.",
    "story": "Every garment packs itself in true stop-motion pops. No human hands. Small pieces jump first, then mid layers fold in two or three frame-steps, then bulk pieces stack. The suitcase fills until the lid slams in three hard bites. The DESERT INN notepad flips to CHECKOUT 11AM.",
    "rhythm_structure": {
      "0-2s": "Locked wide. ROOM 214. Open navy suitcase. Vacation clothes laid out on bed, chair, floor. Subtle frame flicker already on.",
      "2-6s": "Smalls first, discrete jumps with hard shadows: white athletic socks pop-pop into the case; navy boxers fold once and hop in; faded band t-shirt snaps into a rectangle and drops; brown leather belt coils and lands on top.",
      "6-10s": "Mids in jerky folds: light-blue oxford shirt buttons itself in two frame pops then folds; dark-wash jeans fold in three hard steps and slide in; mustard knit sweater compresses and stacks. Notepad pages flip: blank → CHECKOUT 11AM handwritten → hold.",
      "10-13s": "Bulk last: olive windbreaker collapses in two bites and drops in; baseball cap flips onto the pile; leftover sock and undershirt jump from the chair. Case now visibly full.",
      "13-15s": "Lid closes in three stop-frame slams. Final still: closed navy case, brass ROOM 214, notepad frozen on CHECKOUT 11AM. End."
    }
  },
  "camera": {
    "shot_type": "One 15s locked wide of the motel room, 16:9",
    "lens_language": "Still-tripod 32mm, harsh blind stripes across bed and case",
    "camera_journey": "Locked. Only objects change position between frames.",
    "forbidden_moves": [
      "No smooth continuous motion of objects",
      "No hands, arms, or body in frame",
      "No claymation characters",
      "No camera push, pan, tilt, or handheld"
    ]
  },
  "grade": {
    "stock": "Sun-faded motel print, slight strobe between frames",
    "palette": [
      "Desert ochre walls",
      "Brass 214",
      "Navy suitcase",
      "White notepad",
      "Faded cotton, denim indigo, mustard knit, olive nylon"
    ],
    "texture": "Carpet nap, brass, paper, cotton knit, denim, leather belt, nylon shell"
  },
  "audio": {
    "native": true,
    "bed": "Highway hush, AC rattle",
    "accents": [
      "Hard per-frame fabric ticks",
      "Denim fold thumps",
      "Belt coil tap",
      "Paper flip",
      "Lid slam in three hits"
    ],
    "forbidden": [
      "No cartoon boings",
      "No music",
      "No voices"
    ]
  },
  "constraints": [
    "Readable text only: ROOM 214, DESERT INN / FLAGSTAFF, CHECKOUT 11AM",
    "True stop-motion jitter: objects teleport a few centimeters per frame, never float or slide",
    "Packing order visible: socks + underwear + t-shirt + belt, then oxford + jeans + sweater, then windbreaker + cap",
    "Suitcase must look full before the lid closes"
  ]
}

Original post ↗

Storybook girl and rabbit that loop back to the artwork

Shot-by-shot timeline · MiniMax H3 per the author · Aug 29, 2026

The prompt starts and ends on the reference illustration, so the last frame has to rebuild the first. A causality chain (blink, zoom, braid swing, toy, paint splash, type, whip) gives every transition a trigger, and each shot carries its own sound effects, with a closed text ledger of three phrases and an outline rule.

Animated styles and art directionKinetic type and title sequencesCamera moves and transitions

Read @airina_xyz's prompt
15-second, 9:16 surreal storybook-grotesque kinetic animation, preserving the exact hand-painted illustration style, proportions, textures, colors, facial features, clothing, imperfect brushwork, paper-like grain, cream background, and unsettling handmade toy aesthetic of @[image1]. The angry braided girl remains the dominant character; the tiny pink rabbit-doll stays beside her. Maintain character design and spatial continuity throughout. Energetic beat-synchronized editing, exaggerated squash-and-stretch limited to the illustration style, layered 2.5D depth, no photorealism.
TEXT LEDGER — only readable text:“STAY CUTE.”“STAY WEIRD.”“NEVER BEHAVE.”
Typography: chunky hand-painted children’s-book lettering, slightly crooked. Cream-colored letters must have dark red/black outlines; any text matching its background always receives a strong contrasting outline.
VISUAL CAUSALITY: Every transition is triggered by something physically moving in the previous shot: blink → snap zoom → braid swing → toy movement → paint splash → typography → camera whip → seamless restoration.
TIMESTAMPED CHOREOGRAPHY — ~8 SHOTS
0:00–0:01.7 — SHOT 1 — ORIGINAL FRAME Begin exactly on @[image1], perfectly matching the supplied composition. After a brief beat, the girl’s enormous blue eyes snap toward camera; one braid twitches. The rabbit-doll slowly turns its eyes upward.SFX: tiny cloth creak, eyelid tick, faint paper rustle. Music begins with quirky percussion exactly on the eye movement.
0:01.7–0:03.2 — SHOT 2 Beat-hit crash zoom into the girl’s furious face. Her freckles, glossy cheeks and crooked teeth remain painterly. Eyebrows slam downward while both braids whip outward like ropes. Painted red speckles burst behind her.Text punches in: “STAY CUTE.”SFX: braid FWIP, wooden tok, wet paint splut.
0:03.2–0:04.8 — SHOT 3 Camera spirals downward around her overalls while embroidered shapes and tiny painted patches briefly hop off the fabric and orbit her body, then slap back into place on successive beats.SFX: fabric brushing, tiny ceramic clicks, sticker-like pap-pap-pap.
0:04.8–0:06.4 — SHOT 4 The pink rabbit suddenly scampers in a frantic circle around her oversized sneakers, its tiny chain dragging behind it. The girl stomps once; the impact sends a flat illustrated shockwave across the cream floor.Text stretches across the shockwave: “STAY WEIRD.”SFX: rapid plush footsteps, chain jingles, enormous soft THUMP.
0:06.4–0:08.1 — SHOT 5 On the next musical beat, the little stacked object dangling from her left hand telescopes upward like an impossible accordion tower. Camera rapidly cranes upward following it as painted pieces wobble, rotate and clatter. The girl watches with suspicious side-eye.SFX: wooden rattles, tiny crunchy clicks, springy boing.
0:08.1–0:10.0 — SHOT 6 The tower collapses downward in a cascade. Each impact releases hand-painted stars, hearts, scraps and scribbles that swirl violently around girl and rabbit. Her braids spin in the airflow while the rabbit grips its dress. Music percussion accelerates with every impact.SFX: cascading clack-clack-clack, paper flutter, cloth flapping.
0:10.0–0:12.1 — SHOT 7 A red painted ribbon whips across lens, revealing an extreme low-angle view. The girl plants one huge sneaker toward camera, glaring triumphantly; the tiny rabbit imitates her pose beside it.Text lands one word per beat: “NEVER / BEHAVE.”Letters bounce, wobble, then crack slightly like dried paint.SFX: sneaker squeak, paint scrape, tiny rabbit squeak, crunchy letter impacts.
0:12.1–0:15.0 — SHOT 8 — PERFECT LOOP Camera whip-orbits around both characters as stars, scraps, braids, chain and clothing move rapidly in opposing layers. The orbit progressively collapses back into the original frontal composition. Every moving element snaps precisely into its original location—the tower, hands, braids, rabbit, shoes, clothing, expressions and cream background—until the final frame becomes exactly @[image1], pixel-compositionally matching Shot 1 for a seamless infinite loop. Final musical beat lands exactly on the restored frame.SFX: reverse paper flutter → soft whump → tiny final chain ting.
FOLEY + MUSIC SYNC LAW: Background music is playful, eccentric, slightly mischievous percussion with toy-like rhythmic accents; every cut, stomp, zoom, text impact, braid swing and object collision lands on a musical beat. ASMR remains crisp and foregrounded.
CAMERA / DEPTH: rapid push-ins, low-angle exaggeration, spiral track, crane, whip orbit, 2.5D parallax. Preserve screen geography and consistent left/right placement. No unexplained character teleportation.
FAILURE CONTROLS: exact source-art style only; no realism, 3D CGI, anime conversion or style drift; no extra limbs/fingers; no character duplication; no wardrobe changes; no altered facial identity; no malformed rabbit; no random objects; no unreadable text; no camera-side reversals; final frame must precisely reproduce @[image1].

Original post ↗

Engraved garden that opens a hidden path

Shot-by-shot timeline · MiniMax H3 per the author · Aug 28, 2026

Every transition has to start from something already visible: a strand of hair becomes a cypress, plants bend to uncover a path, a shaft of light becomes a highlight on her cheek. A spatial rule keeps every camera on one side of the action axis, the sound is foley only, and a final wipe rebuilds the reference composition.

Animated styles and art directionCamera moves and transitionsKinetic type and title sequences

Read @airina_xyz's prompt
15-second, 9:16 surreal illustrated motion-design film based on @[image1], preserving the exact visual language of the reference: exquisite etched linework, graphic-novel engraving detail, muted dark teal/black vegetation and clothing, luminous coral-salmon sky, architectural surrealism, elegant elongated cypress trees, delicate facial rendering, subtle print texture, dreamlike Renaissance-garden atmosphere. Never become photorealistic, 3D, anime, or painterly.
SUBJECT / WORLD: The young woman remains the visual anchor. She moves through the impossible garden embedded around her silhouette as architecture, vegetation, pathways and her own body briefly interact through elegant graphic transitions. Preserve her facial identity, hairstyle, dark clothing, proportions and restrained expression.
TEXT LEDGER — only readable text:
“SOME PATHS”
“ONLY APPEAR”
“WHEN YOU TURN AROUND.”
Typography: refined narrow serif capitals, pale ivory. Whenever letters cross pale/coral areas, add a thin dark-teal outline; never sacrifice readability.
VISUAL CAUSALITY: Every transition must originate from an object already visible—hair → cypress branches → pathway → architecture → shadow → portrait. No random morphing.
TIMESTAMPED CHOREOGRAPHY — ~8 SHOTS
0:00–0:01.8 — SHOT 1 — REFERENCE OPEN Begin on @[image1] exactly as shown. Extremely subtle 2.5D depth: foreground flowers drift slightly against the woman and distant architecture. A faint breeze catches three loose hairs. Slow push toward her eye.ASMR: soft leaves brushing, tiny hair rustle, distant stone-room ambience.
0:01.8–0:03.5 — SHOT 2 Match-cut one loose strand of hair into the tip of a tall cypress. Camera glides forward down the garden corridor, always staying on the same established side of the woman’s eyeline. Cypress trees streak past in layered parallax.Text materializes between trees: “SOME PATHS”ASMR: crisp leaf whispers, gentle cloth-like swishes.
0:03.5–0:05.2 — SHOT 3 Low tracking shot over dense orange-tipped plants. Their stems bend sequentially like a wave, revealing a previously hidden stone pathway leading toward the coral-lit domed building.ASMR: delicate stems ticking, gravel granules shifting.
0:05.2–0:07.0 — SHOT 4 Camera races gently along that path toward the dome; columns stretch upward in dramatic engraved perspective. A dark doorway opens as if pushed by an unseen breeze.Text appears inside the glowing doorway: “ONLY APPEAR”ASMR: stone scrape, hollow architectural creak, soft air suction.
0:07.0–0:08.7 — SHOT 5 Pass through the doorway into momentary darkness; a narrow shaft of coral light sweeps across frame and becomes the highlight along the woman’s cheek. Match cut to a tight profile close-up. She slowly turns her eyes toward camera before turning her head.ASMR: soft fabric collar movement, single breath, faint skin/hair rustle.
0:08.7–0:10.5 — SHOT 6 As her head turns, her flying hair becomes dozens of thin engraved branches. Camera arcs only within the established 180° hemisphere, revealing the garden now extending behind her. No axis crossing. Cypress silhouettes whip past foreground for speed.ASMR: layered branch swishes, dry ink-like scratching.
0:10.5–0:12.7 — SHOT 7 Wide surreal reveal: she walks slowly away along the newly uncovered path while the towering architecture and cypress forest remain spatially consistent with the original scene. The coral sky glows through the trees.Text assembles from drifting leaf fragments: “WHEN YOU TURN AROUND.”ASMR: soft footsteps on gravel, leaves settling, distant stone resonance.
0:12.7–0:15.0 — SHOT 8 — LOOP-LIKE RETURN She stops and turns back toward camera. Foreground plants rise across lens as a natural wipe; their dark shapes perfectly reconstruct the original composition. End on @[image1] exactly as shown, holding the final frame for ~0.5s.ASMR: close leaf sweep → tiny final rustle → silence.
FOLEY-SYNC LAW: Every visible impact, opening, plant movement, footstep, hair movement and transition receives a precise tactile sound. No music, no dialogue, no voice-over.
CAMERA / SPATIAL LAW: Establish the woman, pathway, dome and garden corridor in shot 1 and preserve those relationships throughout. Maintain consistent screen direction and eyelines. Keep all subsequent cameras within the same 180° side of the action axis; any directional reversal happens visibly through her physical turn, never through an axis-breaking cut.
MOTION DESIGN: engraved-line displacement, restrained parallax, masked botanical wipes, architectural match cuts, silhouette transitions, controlled foreground occlusion, hair-to-cypress transformations, elegant speed ramps.
FAILURE CONTROLS: preserve exact face and hairstyle; no extra characters; no facial distortion; no changing costume; no random buildings; no floating objects; no unreadable or additional text; no excessive morphing; no camera teleportation; no 180° rule violation; no photorealism; no music.

Original post ↗

A life played as a 16-bit game

JSON structure · MiniMax H3 per the author · Aug 28, 2026

One rule runs the piece: each scroll costs visible life points while opportunities appear and vanish. The JSON keeps the protagonist seated while the background scrolls through the years, keeps the phone sprite and life meter on screen throughout, and limits text to game interface elements.

Animated styles and art directionKinetic type and title sequences

Read @airina_xyz's prompt
{
  "archetype": "darkly comic life simulation",
  "duration": "15s",
  "prompt": {
    "concept": {
      "title": "One More Video",
      "description": "A person's entire life plays out like a retro video game while the player chooses 'SCROLL' instead of participating.",
      "duration": "15s",
      "core_visual_rule": "Every scroll consumes visible LIFE points while opportunities appear briefly and permanently disappear.",
      "rhythm_structure": "choice → lost opportunity → faster choices → life bar collapse"
    },
    "subject_lock": "Same pixel-art protagonist, aging through recognizable sprite variations.",
    "typography": "Minimal integrated game UI: LIFE meter, AGE counter and contextual choices only.",
    "camera_direction": "Side-view game framing with occasional dramatic pixel zoom.",
    "visual_style": {
      "overall": "early-1990s 16-bit pixel-art adventure game",
      "medium_or_ontology": "everything built from crisp hand-placed pixels and tiled sprite animation",
      "color_palette": "limited saturated 16-bit palette",
      "materials_and_texture": "pixel dithering, sprite tiles, parallax backgrounds",
      "lighting": "stylized game lighting and palette swaps",
      "depth_and_composition": "side-scrolling layered environments"
    },
    "motion_language": {
      "core_behavior": "scrolling triggers a UI click, drains LIFE and advances the environment without moving the protagonist",
      "transition_system": "screen scrolls horizontally through years while character remains seated",
      "continuity_rule": "phone sprite and LIFE meter always visible",
      "motion_intensity": "slow opening, accelerating time skips"
    },
    "storyboard": {
      "total_beats": 5,
      "beat_duration": "3s",
      "beats": [
        {"id":1,"time":"00:00-00:03","description":"AGE 18. Hero sits on bed. Window shows friends outside. Choices appear: GO OUT / SCROLL. Cursor selects SCROLL.","subject_action":"scrolls","camera":"classic side view","transition":"LIFE meter drops slightly"},
        {"id":2,"time":"00:03-00:06","description":"AGE 25. Choices: CREATE SOMETHING / SCROLL. SCROLL is selected again. Art supplies vanish from the room.","subject_action":"scrolls","camera":"side view","transition":"room tile set changes"},
        {"id":3,"time":"00:06-00:09","description":"AGE 38. Travel train waits outside. Choices flash briefly; SCROLL wins automatically.","subject_action":"rapid scrolling","camera":"pixel zoom","transition":"train departs and AGE counter jumps"},
        {"id":4,"time":"00:09-00:12","description":"AGE counter races 45…58…67…79 while seasons flicker outside. LIFE bar reaches one pixel.","subject_action":"thumb loops endlessly","camera":"locked","transition":"increasingly rapid palette swaps"},
        {"id":5,"time":"00:12-00:15","description":"AGE 82. LIFE reaches zero. Game screen freezes. A final menu appears: TOTAL SCREEN TIME: 21 YEARS.","subject_action":"phone slips from hand","camera":"slow pixel zoom-out","transition":"GAME OVER text appears"}
      ]
    },
    "end_frame": "Black pixel screen with tiny text: 'GAME OVER — Continue scrolling?' and a blinking cursor hovering over YES.",
    "negative_prompt": "3D voxel art, modern high-resolution game graphics, photorealism, smooth vector edges, realistic social-media logos, complex UI clutter"
  }
}

Original post ↗

Tilt-shift UFO invasion of a toy city

Shot-by-shot timeline · MiniMax H3 per the author · Aug 24, 2026

The style block defines the miniature look in optical terms: a narrow focus band, heavy blur, toy colours and a high angle. Four hard-cut acts each carry their own audio line, and the last one pulls up to reveal the whole tabletop, which pays off the style.

Animated styles and art directionGiants, armies and epic scale

Read @MayorKingAI's prompt
Tilt-shift miniature photography style, 15 seconds, 16:9. Everything looks like a tiny handcrafted diorama — extreme shallow depth of field with sharp focus on a narrow band and heavy blur above and below, oversaturated toy-like colors, high angle looking down on a miniature city. The toy-model look contrasts with chaotic events.

ACT 1 (0-4s): High-angle view of a bustling sunny miniature city, alive with motion — tiny cars flowing down streets, little figures walking briskly along sidewalks, toy buses stopping and going. Suddenly classic silver flying saucers — round metallic discs with domed tops and spinning colored lights underneath — descend from the clouds, casting shadows over the skyline. Traffic slows as shadows fall. Audio: city ambience, tiny car horns, gentle hum building into a deep alien drone.

ACT 2 (4-8s): Hard cut. The flying saucers fire glowing beams down; a miniature skyscraper crumbles into tiny falling blocks. Little figures scatter and run through the streets like panicked ants, tiny cars crashing. Dust rises. Slight camera shake. Audio: crumbling rubble, tiny distant screams, blaring car alarms, energy beam blasts.

ACT 3 (8-12s): Hard cut. Glowing tractor beams from the saucers lift tiny people and toy cars up into the sky — miniature abductions, figures rising helplessly toward the domed ships. Beams sweep across the diorama city. Audio: rising sci-fi whir of tractor beams, faint terrified screams, chaos.

ACT 4 (12-15s): Hard cut. Wide high-angle of the whole miniature city in chaos — fires like tiny orange dots, crumbling toy towers, silver saucers hovering above with spinning lights. The camera pulls up and the entire scene looks like a tabletop model. Audio: distant sirens, deep rumble, one final ominous alien drone.

No dialogue, no voice.

Original post ↗

Watercolour mystery: the queen or the prince

Shot-by-shot timeline · MiniMax H3 per the author · Aug 24, 2026

Transitions are ink bleeds native to the medium, and colour carries the plot: gold for the kingdom, blue for night, and a red stain of guilt that moves from the girl to the prince's hands. The audio line sets instruments and brush sounds, with no voice.

Animated styles and art directionTransformation and style shifts

Read @MayorKingAI's prompt
Watercolor ink animation, 15 seconds, 16:9. Fluid translucent watercolor style — bleeding ink washes on textured paper, silhouetted figures in gold, deep red and black, edges dissolving and reforming, scenes bleeding softly into one another with flowing ink transitions. No voice, audio: soft cello and harp, water and brush sounds, a low narrator's hum.

ACT 1 (0-4s): A watercolor kingdom blooms into being — golden washes forming a castle on a hill, a young prince and a farmer's daughter as silhouettes walking together. Warm gold tones. Camera drifts slowly across the wet paper.

ACT 2 (4-8s): Bleed transition. Night falls in deep blue washes. The two figures sleep beneath a great yew tree. At dawn the girl lies still — a spreading red stain blooms across the paper beneath her. The prince's silhouette rises alone.

ACT 3 (8-12s): Bleed transition. The prince points an accusing hand toward a distant tower where a queen's silhouette stands. Red washes surge — a mob of shadow figures with torches floods across the paper toward her. Ambiguity: was it her, or him?

ACT 4 (12-15s): The queen's silhouette dissolves into scattering ink as the crowd overtakes the tower. The prince stands crowned against a calm golden sky. He looks down at his own hands — a dark red stain slowly blooms across his palms. Everything dissolves into one spreading red drop on white paper. Final cello note.

Original post ↗

Pop-up book dragon

Shot-by-shot timeline · MiniMax H3 per the author · Aug 22, 2026

Every event is a paper mechanism: pages rise, stalactites unfold and the dragon opens in layered sheets. A rumble under the music box grows across the four acts and pays off in the final roar.

Animated styles and art directionGiants, armies and epic scale

Read @MayorKingAI's prompt
Pop-up book paper craft animation, 15 seconds, 16:9. Layered cut-paper 3D pop-up book style, visible paper folds and tabs, soft warm storybook lighting, hand-crafted texture, shallow depth of field on an open book. No voice, audio: gentle music box, page-turning sounds, a low rumble building to a roar.

ACT 1 (0-4s): An open storybook on a wooden table. The first paper page rises to reveal a layered cut-paper mountain landscape at dusk, tiny folded pine trees standing up. Slow push-in toward a dark cave opening in the paper mountain.

ACT 2 (4-8s): Hard cut. A page turns and pops up a cave interior — layers of paper stalactites unfold, and a pile of folded golden paper treasure rises into place, glittering. Camera drifts deeper into the cave.

ACT 3 (8-12s): Hard cut. Among the paper treasure, a massive folded paper dragon slowly unfolds and rises — wings spreading in layered paper sheets, neck extending upward, filling the frame. The rumble grows.

ACT 4 (12-15s): Hard cut. Close-up as the paper dragon's eye snaps open, glowing amber, and it lets out a roar — paper flames unfold from its jaws toward the camera. Freeze on the dragon filling the whole open book. Final roar.

Original post ↗

Three Kingdoms title sequence in Chinese poster style

Continuous action chain · MiniMax H3 per the author · Aug 22, 2026

Each reference image becomes a title card assembled from ink, torn silk, bronze patterns and smoke instead of being cut in. A surname in calligraphy appears every two images, and the last card holds longer for the closing title.

Animated styles and art directionKinetic type and title sequencesReference-driven characters

Kitsune transformation in woodblock style

Shot-by-shot timeline · MiniMax H3 per the author · Aug 21, 2026

The hook sits in the first second, an eye snapping open, before the pull-back. The transformation is mapped part to part (the tails blur into kimono sleeves) while the ears and tails remain, so the creature stays readable in its new form.

Animated styles and art directionTransformation and style shifts

Read @MayorKingAI's prompt
Ukiyo-e Japanese woodblock print animation, 15 seconds, 16:9. Flat color blocks, bold black outlines, visible washi paper grain, muted indigo/vermilion/gold palette, stylized clouds and huge full moon, subtle woodblock registration texture. No voice, audio: koto strings, rising wind, a single haunting fox cry.

ACT 1 (0-3s): Immediate hook — extreme close-up of a white fox's eye snapping open, gold iris, blue kitsune-bi flame reflected in it. Camera pulls back to reveal a nine-tailed fox on a moonlit hilltop, all nine tails fanning out dramatically. Floating blue flames drift around it.

ACT 2 (3-9s): The fox rises onto its hind legs and begins to transform — body wreathed in swirling blue flames and mist, silhouette shifting upward, tails blurring into flowing kimono sleeves. Slow vertical rising motion, flames intensifying.

ACT 3 (9-13s): Transformation completes — an elegant woman in a flowing kimono, fox ears and nine tails still visible behind her, standing among the blue flames. Her sleeves and hair ripple in the wind. Petals swirl past. Gentle push-in.

ACT 4 (13-15s): She turns sharply to face the viewer; her eyes flash brilliant gold and a fox shadow flickers behind her. She smiles. Freeze on her face as flames flare around the frame. Final sharp koto note.

Original post ↗

Thermopylae as red-figure pottery

Shot-by-shot timeline · MiniMax H3 per the author · Aug 20, 2026

Flat silhouettes in profile keep a battle readable in a medium without depth. Four hard-cut acts move from phalanx to arrows to clash to last stand, and the final pull-back reveals the scene painted on a turning amphora.

Animated styles and art directionClose combat and impactGiants, armies and epic scale

Read @MayorKingAI's prompt
Ancient Greek red-figure pottery animation, 15 seconds, 16:9. Flat 2D style: matte black silhouetted warriors with fine terracotta line detail, solid terracotta-orange background, Greek meander border framing top and bottom, subtle glaze crackle texture. No voice, audio: epic war drums, bronze clashes, whistling arrows.

ACT 1 (0-4s): Narrow cliff pass. Leonidas and Spartan hoplites lock shields into a phalanx wall, lambda emblems visible, spears rising in unison. Slow lateral tracking shot.

ACT 2 (4-8s): Hard cut. Thousands of Persian arrows fly as black streaks across the terracotta sky, blotting it out. Spartans crouch under shields, arrows shattering on bronze. Low angle.

ACT 3 (8-12s): Hard cut. Frontal clash — Persian warriors crash against the shield wall, Spartan spears thrust over the rims, figures tangle in stylized profile combat. Dynamic push-in.

ACT 4 (12-15s): Hard cut. Leonidas stands alone among fallen warriors, spear raised. Camera pulls back revealing the entire scene is painted on a rotating ceramic amphora against a black museum background. Final drum hit.

Original post ↗

The fall of Troy in paper cut-out stop-motion

Shot-by-shot timeline · MiniMax H3 per the author · Aug 19, 2026

The style block bans smooth camera moves and dissolves, so every change is a hard stop-motion jump, including day turning to night by swapping paper layers. The horse is tracked as one object through all four beats and ends standing in the burning square.

Animated styles and art directionClose combat and impactGiants, armies and epic scale

Read @MayorKingAI's prompt
STYLE: Handcrafted paper cut-out stop-motion diorama. Cardboard buildings, ink-silhouette figures, torn paper edges, visible fibers, soft lamp shadows. Locked wide shots, hard stop-motion jumps. No smooth camera moves, no dissolves, no digital effects.

PALETTE: Ochre, terracotta, turquoise, bronze gold, charcoal, ember orange.

AUDIO: No voice-over. Wind, creaking wooden wheels, rope tension, war drums, crackling fire.

STORY — The Fall of Troy, 15s:

0–3.5s — Wide shot: the gates of the walled paper city of Troy open. A giant corrugated-cardboard horse standing on four wooden wheels is pulled inside by dozens of tiny Trojan silhouettes hauling long paper ropes. The horse rolls forward frame by frame through the gates.

3.5–7.5s — Hard jump: the horse now sits in the center of the main square, ropes dropped on the ground, Trojans dispersing. The sky shifts to deep navy night in a single stepped transition — day paper layers swapped for night layers.

7.5–11.5s — Hard jump: a hatch in the horse's belly swings open and a cardboard ramp slides down to the ground. Greek soldier silhouettes march down the ramp in a line, jerky frame-by-frame, and clash with Trojan soldiers in the square — spears and shields colliding in stepped bursts as the first torn orange paper flames appear on rooftops.

11.5–15s — Hard jump to a wide exterior shot: the whole city burns with layered paper flames and rising charcoal smoke strips against the night sky. Final static hold on the silhouette of the horse still standing in the burning square.

Original post ↗

Where it shows up alongside other effects

Anime horror-night gag with burned-in subtitles

Shot list · MiniMax H3 per the author · Oct 6, 2026

Six shots have exact cut times, a field of view and a static camera each; the Japanese lines are spoken, and the English subtitles are specified as image-only text with typeface, outline, a single line, a fixed position and a rule that each one leaves at the cut. It reverses the usual subtitle ban on the same principle: every piece of on-screen text is listed exactly, and shots without a line are told to carry none. Gag faces are written as concrete anime conventions (shrinking pupils, shade lines, souls leaving the mouth), with a closing rule that faces return to the normal design.

Dialogue and native soundPerformance and emotionAnimated styles and art direction

Read @MayorKingAI's prompt
2D anime | romantic comedy | cel-shaded, clean line art, dark blue night palette.
Multishot, 6 shots. Cuts only at the specified points.

CHARACTERS
HANA: wavy shoulder-length dark red hair, oversized gray cardigan with one button missing, a bandage on her index finger. 160 cm.
YUTO: neat black hair with a small bed-head tuft at the back, navy hoodie with a pilled sleeve, glasses pushed up on his forehead. 178 cm.

LOCATION
Dark living room at night. They share one sofa under a single fleece blanket. A TV off-frame in front of them is the only light source, flickering blue on their faces. A bowl of popcorn between them.

SUBTITLES
Spoken dialogue is in Japanese. English subtitles are burned into the image as on-screen text only — they are never spoken aloud. Style: white sans-serif, thin black outline, one single line, bottom center of frame, 8% from the bottom edge. Each subtitle appears with its line and disappears at the cut. Subtitle text is shown inside 【】 brackets below; the brackets themselves are not displayed.

Shot 1 (0–2.5s) — Medium front shot of both, FOV 47, static.
Hana sits upright, arms crossed, smug smile, chin raised. Yuto beside her grips a cushion with both hands, already tense.
Hana (spoken, Japanese): 「全然怖くないし。」
On-screen subtitle: 【I'm not scared at all.】
HARD CUT

Shot 2 (2.5–5s) — Close-up on Yuto, FOV 29, slow push-in ending on his eyes.
His pupils shrink to tiny dots, a vertical column of blue shade lines drops over his upper face, sweat drop on his temple. He swallows, the cushion squeezed flat. No subtitle in this shot.
HARD CUT

Shot 3 (5–8s) — Medium front shot, FOV 47, static.
A violent white flash from the TV and a loud sting. Both jump. Clear gag faces: mouths huge and wavy, eyes white, hair standing in spikes. The popcorn bowl flips and popcorn bursts upward like confetti. Two small white ghost-shaped souls slip out of their open mouths.
Hana and Yuto (spoken, Japanese): 「ぎゃああああっ!」
On-screen subtitle: 【AAAAAAH!】
HARD CUT

Shot 4 (8–10.5s) — Wide shot, FOV 63, static.
They are clutching each other tightly, trembling, their outlines drawn with shaky wobbling lines. Popcorn rains down on them. The souls drift back into their mouths. No subtitle in this shot.
HARD CUT

Shot 5 (10.5–13s) — Two-shot close-up, profile, FOV 29, static.
Silence. They slowly realize how close they are, faces a hand apart. Both blush deep red at the same time. A tiny comic sparkle pops between them. They let go and shuffle apart in short jerky movements. No subtitle in this shot.
HARD CUT

Shot 6 (13–15s) — Close-up on Hana, three-quarter front, FOV 29, static.
Normal detailed anime face: cheeks red, eyes sliding sideways, lips pouting. One popcorn kernel stuck in her hair.
Hana (spoken, Japanese): 「…今のはノーカン。」
On-screen subtitle: 【...That one doesn't count.】
She pulls the blanket up over her nose. End on her eyes peeking above the blanket.

LIGHTING
Only light source is the off-frame TV: flickering cool blue 8500K, with a single white flash in shot 3. Deep shadows in the room.

Sound: No score. Production audio only — faint TV dialogue murmur, a sharp horror sting, popcorn scattering, fabric rustle, double scream, then near silence and a soft blanket rustle. All voices in Japanese only.

Characters keep the same face, hair and wardrobe in every shot. Gag faces are clear, exaggerated super-deformed expressions, then return to the normal detailed anime face. Clean consistent line art, consistent lighting between shots. Subtitles stay in the same position, size and style in every shot, spelled exactly as written. The room contains only the sofa, the blanket and the popcorn bowl.

Original post ↗

A hand-drawn tiny baker in a live-action kitchen

sections · MiniMax H3 per the author · Sep 30, 2026

The scale is built into every beat: each one pairs the small drawn baker with a full-size object he has to deal with (a flour bag, a stick of butter, a mixer attachment), so there is always something to measure him against. The style is stated as a hybrid of a hand-drawn character and a photoreal kitchen, and the camera list favours macro and low angles that keep the size contrast in frame. The beats carry no times, so pacing is left to the model; the post credits the concept to @AIwithSynthia.

Giants, armies and epic scaleAnimated styles and art direction

Read @airina_xyz's prompt
Subject
A miniature hand-drawn 2D animated boy with messy brown hair, round glasses, a blue apron, striped socks, and a tiny silicone spatula, baking cookies inside a realistic full-sized kitchen.
Style
Mixed-reality hybrid animation combining playful 2D hand-drawn doodle character art with a photorealistic live-action kitchen. Whimsical stop-motion physics, expressive line animation, exaggerated scale, warm cinematic lighting, and playful cooking sound effects.
Camera & Framing
Macro close-ups, low-angle tracking shots, overhead kitchen perspectives, dramatic push-ins, and shallow depth of field emphasizing the tiny baker against oversized kitchen objects.
Audio & Scene Breakdown
Visual: The tiny baker stands beside a giant bag of flour, trying to push it open with his little spatula. A burst of flour explodes around him like a snowstorm.
Visual: He scrambles onto the edge of a huge butter dish and slides down a stick of butter into a giant mixing bowl.
Visual: Inside the bowl, he dodges chunks of chocolate chips falling from above like boulders while cookie dough is being stirred around him.
Visual: He grabs onto a massive electric hand mixer attachment and spins in circles as the dough swirls beneath him.
Visual: He leaps out onto a baking tray and uses an oversized spoon to shape cookie dough blobs, getting comically stuck in one of them.
Visual: He rides a rolling pin like a balancing beam while trying to flatten one runaway cookie.
Visual: The camera pulls back to reveal a tray of freshly baked cookies. The tiny baker proudly places one chocolate chip on top of the biggest cookie—then the cookie suddenly cracks and tilts, sending him sliding off.
Audio: Playful kitchen ambience, flour poofs, dough squishes, mixer whirring, tray clinks, tiny footsteps, comedic impact sounds, subtle upbeat music, no dialogue.

Original post ↗

A 24-cut sword duel written as rules, not a shot list

brief · MiniMax H3 per the author · Sep 21, 2026

Instead of listing the 24 cuts, the prompt sets rules for every exchange: one fighter attacks with force and re-entry, the other answers with footwork and redirection, and every reversal must follow clear contact and consequence. Camera angles are given jobs rather than times (low tracking for footwork, overhead for attack geometry), and the loose ink style limits effects to velocity strokes and selective impact flashes. It is told to cut out mid-exchange with both fighters still attacking, so no outcome has to be staged.

Close combat and impactAnimated styles and art directionCamera moves and transitions

Read @aimikoda's prompt
24 fast cuts. Loose hand-drawn anime: thin broken ink, rough painterly fills, elongated anatomy, dark negative space and clean katana glow. On the first frame, one fighter drops from above into the other’s intercepting blade. One attacks through force bursts and relentless re-entry; the other counters with triangular footwork, angular evasion and redirection. Walls, ledges, columns, platforms and open drops continually reshape their exchanges. Missed strikes redirect midair; brief blade binds become angle steals, hilt traps and one-handed switches; near-misses slice hair and sleeves. Clashes feel countless, but every reversal follows clear contact and consequence. Low tracking reveals footwork, overhead shots expose attack geometry, while blade-line pursuit, vertical whip climbs, falling top-down views, near-POV passes, obstruction reveals, lateral rushes and hard reframing follow changing advantages. Scratchy velocity strokes, directional trails, hair arcs and selective impact flashes keep movement readable without overwhelming effects. Ignition snaps, fast pass-bys, pitch-bending swings and dense metallic clashes form rhythmic percussion. Cut mid-exchange with both still attacking and intensity rising.

Original post ↗

Restyling a grey type animation through a reference video

Shot-by-shot timeline · MiniMax H3 per the author · Sep 14, 2026

The post says the grey source animation was built in Blender with another model, and the H3 prompt makes that video the sole reference for type geometry, motion, camera and editing, leaving the text to handle materials, light and sound. The six words are listed in order with a 2.5-second section each, and every section gets its own palette, material and sound effects tied to motions already in the source. It is a clear split of jobs between a reference video and the written prompt.

Kinetic type and title sequencesReference-driven charactersAnimated styles and art direction

A character speedpaint driven by a visible cursor

Shot-by-shot timeline · MiniMax H3 per the author · Sep 12, 2026

Every stage has a cause on screen: a cursor drives each action, and the prompt says twice that nothing may appear on its own and that each stage grows out of the one before. Five timed stages (guides, sketch, line art, flat colour, shading) lead to a reveal that must closely match the reference, so the reference image works as the end state rather than only as a style guide. Small undo-and-redo strokes are written in to keep the process looking human.

Transformation and style shiftsReference-driven charactersAnimated styles and art direction

Read @AIwithJessica's prompt
Video Prompt:
Aspect Ratio: 16:9 | Duration: 15 seconds
Title: Digital Character Build — Blue-Haired Cyber Idol Speedpaint

Prompt:

Capture a fast, energetic 15-second screen-recording style timelapse showing an artist building an anime character from scratch inside a modern digital painting program, framed in widescreen 16:9. Keep a visible cursor driving every action throughout — nothing should pop into existence on its own; each stage must visibly grow out of the one before it.

0:00–0:02 — Empty Canvas, First Marks

Open on a bare white workspace inside a sleek illustration app. The cursor darts across the screen laying down rough guide shapes — a basic head outline, shoulder line, and loose gesture for a half-body female portrait pose.
0:02–0:05 — Building the Sketch

Pace picks up as the cursor fleshes out facial structure, layered twin-tail hairstyle with twin topknots, front-swept bangs, and a rough outline of a fitted sleeveless mock-neck top. Small pauses for redo/undo strokes keep it grounded and human-feeling.
0:05–0:08 — Ink Pass / Clean Lines

The messy sketch tightens into crisp linework. Cursor toggles between brush sizes, sharpens eye shape, mouth, dangling earrings, stacked necklaces, and a chunky wrist chain. Guide sketches fade out of view.
0:08–0:11 — Flat Color Blocking

Speedy color-fill segment: pale aqua hair, soft ivory skin, vivid blue irises, deep black glossy garment, warm gold-toned accessories, and pink floral hair clips. Show the cursor swapping palettes, sampling colors, and using mask layers to fill precisely.
0:11–0:13 — Light, Shadow & Texture Pass

Cursor lays in subtle shading beneath the hairline and chin, adds glassy shine to eyes and lips, gives the gold jewelry a metallic gleam, and adds soft fabric creases to the top. Push in for a close-up on the face and accessories mid-detail, then pull back out.
0:13–0:15 — Reveal

Toolbars and sketch guides disappear from view. The finished portrait now closely mirrors the reference art — icy blue twin-tails with rose-covered buns, sharp blue-eyed gaze, glossy black top, gold jewelry, all set against a striking black-and-crimson angular backdrop. One last minor cursor tweak, then the frame holds steady on the completed piece.
Style: Crisp digital-timelapse aesthetic, believable creative-software interface, fluid and legible cursor movement, brisk but trackable pacing, clean panel and layer transitions, no on-screen text, no invented dialogue, no random pop-ups.

Motion: Maintain constant forward momentum for all 15 seconds — no dead air. Condense a long real-world drawing session into a tight, readable sequence: blank page → loose sketch → refined linework → base colors → shading/polish → final piece.

Camera: Treat this as a screen-capture with light digital zoom/pan following the cursor and active work area. Interface stays steady and legible — no cinematic camera cuts breaking the software frame.

Note: The character should never materialize in one instant jump — every phase must show clear cursor-driven progress building on the last.

Original post ↗

Cola parody where every pour drains the oasis

JSON structure · MiniMax H3 per the author · Aug 29, 2026

A JSON concept states one rule (each pour removes a matching amount of water from the landscape) and splits 15 seconds into miracle, exchange and punchline. The bottle and glass stay photoreal while the world turns to watercolour, and the negative prompt limits the satire and puts a parody label on the end card.

Product, cars and trailersTransformation and style shiftsAnimated styles and art direction

Read @airina_xyz's prompt
{
  "archetype": "poetic anti-ad / environmental satire",
  "duration": "15 seconds",
  "prompt": {
    "concept": {
      "title": "TASTE THE REFRESHMENT",
      "description": "A magical Coca-Cola oasis commercial slowly reveals that the bottle's abundance is being borrowed from the landscape around it.",
      "duration": "15s",
      "core_visual_rule": "Every milliliter poured into the glass removes an equal-looking visual quantity of water from the surrounding world.",
      "rhythm_structure": "luxury miracle 0-5s → impossible exchange 5-11s → bleak visual joke 11-15s"
    },
    "camera_direction": "Begin like a high-budget perfume/beverage campaign with floating macro droplets and graceful dolly motion. Transition into locked wide compositions as the environment dries.",
    "visual_style": {
      "overall": "photoreal luxury desert commercial dissolving into delicate watercolor-and-ink animation",
      "color_palette": "deep Coke red, gold sunlight, crystalline blue shifting into ochre and faded beige",
      "materials_and_texture": "condensation becomes watercolor pigment; cracks spread like ink lines",
      "lighting": "radiant sunrise gradually bleaching into dry noon"
    },
    "motion_language": {
      "core_behavior": "pour liquid → landscape water retreats",
      "transition_system": "physical cause-and-effect synchronized exactly with the pour",
      "continuity_rule": "the bottle and glass remain photoreal while the world becomes painted"
    },
    "storyboard": {
      "total_beats": 6,
      "beats": [
        {"time":"00:00-00:02.5","description":"A pristine Coca-Cola bottle sits beside a spectacular blue oasis. Condensation sparkles. Whispered VO: 'Refreshment comes from somewhere.'","camera":"luxury macro push"},
        {"time":"00:02.5-00:05","description":"Beautiful cola begins pouring into an ice-filled glass.","camera":"slow-motion side profile"},
        {"time":"00:05-00:07.5","description":"As the glass fills, the oasis waterline mysteriously drops by the same amount. Palm reflections vanish.","camera":"rack focus from glass to oasis"},
        {"time":"00:07.5-00:10","description":"The pour continues. Nearby irrigation channels retract backward, becoming thin blue watercolor lines disappearing into the bottle.","camera":"wide lateral track"},
        {"time":"00:10-00:12.5","description":"Glass is now perfectly full and sparkling. Behind it: cracked dry earth. The advertising music abruptly stops.","camera":"static hero composition"},
        {"time":"00:12.5-00:15","description":"A single ice cube clinks. Text: 'TASTE THE REFRESHMENT.' Beat. Smaller line appears: 'Everybody else was thirsty too.' PARODY — NOT AN OFFICIAL AD.","camera":"motionless"}
      ]
    },
    "end_frame": "perfect cold glass foreground, visibly depleted painted landscape background",
    "negative_prompt": "no claim that all Coca-Cola plants cause drought, no fabricated numbers, no suffering people, no political slogans, allegorical environmental satire only"
  }
}

Original post ↗

Chibi reaction chain that never returns to normal

Shot-by-shot timeline · MiniMax H3 per the author · Aug 28, 2026

The normal form appears once, then each emotion morphs straight into the next in two-second windows, each with its own animation language and one word of kinetic type. A white studio and a mostly static camera keep the attention on the transformations.

Transformation and style shiftsAnimated styles and art directionKinetic type and title sequences

Read @aimikoda's prompt
Create a 15-second anime character transformation showcase on a pure white seamless studio background. Use one single anime character throughout with perfectly consistent face, hairstyle, outfit, body proportions and colors. Keep the camera mostly static in a clean full-body or 3/4 studio framing.

Show the character in a normal anime form only at the beginning. The character smiles at the camera and gives a small friendly wave. After that, the character must never return to the normal form. Each new emotional state should morph directly into the next one in one continuous transformation chain.

0-1s
Show the character in their normal anime form for the first and only time. They smile naturally at the camera, give a small friendly wave and hold a simple clean pose with subtle breathing.

1-3s
The character suddenly transforms into an exaggerated super-deformed goofy form with loose limbs, asymmetrical eyes, a ridiculous grin and playful bouncing movement. Large kinetic typography appears: "SILLY". The letters wobble and bounce with the character.

3-5s
Without returning to normal, the silly form morphs directly into a tiny shy chibi form. Shoulders raised, knees turned inward, intense blush, nervous fingers and avoiding eye contact. Typography appears: "SHY". The letters softly shrink and squeeze inward around the character.

5-7s
The shy form morphs directly into an extreme shocked reaction. Huge eyes, tiny pupils, wide-open mouth, rigid body and explosive comic speed lines. Typography appears: "SHOCKED!" and rapidly expands outward like a visual explosion.

7-9s
The shocked form transforms directly into a fiery angry chibi form. Clenched fists, puffed cheeks, shaking body, stomping feet and a classic anime anger symbol on the forehead. Typography appears: "ANGRY" and slams heavily into frame.

9-11s
The angry form morphs directly into a frantic tiny chibi version. Arms flailing, sweat drops flying everywhere, legs scrambling rapidly in place and chaotic motion trails surrounding the body. Typography appears: "PANIC!" and shakes, scatters and struggles to stay aligned.

11-13s
The panic form transforms directly into a simplified smug caricature. Half-lidded eyes, tiny confident smile, raised chin and exaggerated proud posture. Typography appears: "SMUG" and slides smoothly into place.

13-15s
The smug form morphs directly into a monochrome devastated chibi form collapsed on the floor. Empty eyes, slumped posture and a tiny ghost-like soul slowly floating out of the body. Typography appears: "DONE." and drops downward, loses energy and deflates beside them. Hold the final defeated pose.

Use sharp snap transitions, squash-and-stretch, proportion morphing, smear frames, comic symbols and expressive kinetic typography. Each emotional form should have its own distinct animation language. Fast, playful, polished and highly readable, like an anime reaction transformation reel.

Original post ↗

Guess-the-anime motion-design template

template · MiniMax H3 per the author · Aug 27, 2026

A reusable template with one variable, the source title, which the prompt bans from appearing in any form. It asks the model to choose spoiler-safe clues that grow more recognisable, to derive palette and type from the source, and to end abruptly on an unresolved image.

Kinetic type and title sequencesAnimated styles and art direction

Read @aimikoda's prompt
Create an intense fast-paced 2D cinematic motion-design sequence based on {ANIME_TITLE}.

Do not show, write or reveal the anime title at any point.
Do not use character names, logos, answer cards, meta text or major plot spoilers.

Build the video as 15 distinct escalating visual beats. Do not label the beats on screen. Infer the 15 strongest spoiler-safe visual clues from the anime and transform them into increasingly recognizable cinematic moments.

The visual direction should feel bold, stylized and premium, combining expressive 2D anime animation with high-end graphic design.

Avoid clean flat-vector explainer aesthetics and avoid 3D CGI.

Use a rich mixed 2D language:
hand-drawn anime linework,
cel-shaded silhouettes,
rough ink strokes,
painted shadows,
screen-print textures,
paper grain,
halftone fragments,
graphic cutouts,
sharp geometric framing,
layered editorial collage,
abstract symbols,
speed lines,
impact frames,
smears,
brush transitions,
high-contrast shape design,
dramatic negative space.

Some clues can appear as full stylized illustrations, while others can be reduced into silhouettes, textures, symbols, fragments or abstract visual metaphors.

Let each anime determine its own visual system. Derive the color palette, line quality, textures, typography, transition language and overall graphic identity from the source material.

Use typography only when visually appropriate and never for explanatory text. Fonts should feel native to the anime’s world and aesthetic.

Keep the sequence extremely energetic and visually surprising. Use rapid composition changes, whip transitions, smash cuts, match cuts, sudden close-ups, fast pans, graphic wipes, layered 2D parallax, frame-breaking motion, aggressive scale changes and rhythmic impact frames.

Avoid repetitive shot structure. Every beat should feel visually different.

Move between:
minimal symbolic frames,
dense graphic compositions,
stylized environmental glimpses,
extreme detail shots,
bold silhouettes,
abstract transformations,
explosive anime-style action imagery.

The clues should begin ambiguous, then progressively become more recognizable to fans without revealing major story information.

Transitions should creatively transform one clue into another through shape, color, line, texture or motion rather than simply cutting between unrelated images.

The final few beats should accelerate dramatically, remixing the strongest visual motifs from earlier moments into a dense, high-energy climax.

End abruptly on a powerful unresolved visual clue.

No title reveal.
No confirmation.
No guessing-game text.

The final result should feel like an experimental premium anime opening sequence, graphic design film and high-energy 2D animation fused together, visually striking enough to work even without knowing the anime.

Original post ↗

City map that builds itself as the camera passes

template · MiniMax H3 per the author · Aug 25, 2026

One variable, the city, drives the piece, and the camera rule ties speed to content: fast between areas, slower at landmarks so they can pop up and settle. The city name is held back until a final title designed to fit local culture.

Camera moves and transitionsKinetic type and title sequencesAnimated styles and art direction

Read @aimikoda's prompt
[CITY] = city name here

Create a stylized 3D animated city-map sequence inspired by [CITY].

Show a beautifully designed miniature city map built on a solid land surface.
Do not show the city name during the map sequence.

Make the entire world feel like a landscape of memories.
Every street, building, park, bridge, stairway, waterfront, and landmark should feel nostalgic, intimate, and emotionally charged.
Blend architecture, pathways, terrain, and natural elements into one compact, cohesive miniature world.

Use one continuous cinematic shot.
The camera moves quickly and smoothly across the map with energetic forward motion, gentle turns, and subtle height changes.
As the camera reaches major iconic locations, briefly slow down so those places can pop up and settle clearly.
Then accelerate again and continue toward the next area.

As the camera advances, buildings, streets, trees, bridges, and landmarks rise from the map in satisfying pop-up motion.
Use quick upward emergence, slight overshoot, soft bounce, and clean settling.
Let each important location pop up in a memorable way, with nearby supporting elements appearing around it.
The city should progressively assemble along the camera path so the world feels like it is awakening through memory.

Show only the most iconic and culturally recognizable elements of [CITY].
Keep the sequence compact, visually rich, and easy to read in 15 seconds.

At the end, pull back slightly and transition to a clean final title moment.
Display the city name prominently in the center of the screen as a typographic title.
Design the typography according to the local culture and visual identity of [CITY].
The lettering should feel stylish, elegant, and culturally grounded rather than generic.
Keep it clearly readable and visually striking as the final reveal.

Use a richly detailed stylized 3D diorama look with appealing materials, warm daylight, gentle shadows, a bright sky gradient, and a polished premium finish.

Original post ↗

One character board, two opposite uses

sections · MiniMax H3 per the author · Aug 22, 2026

Two prompts use the same board in opposite ways. One animates the board in place with a fully static camera and rules for each panel (portrait poses, turntables, small actions inside frames); the other uses it as reference only and builds a cinematic intro, showing how one sentence changes a reference's role.

Reference-driven charactersAnimated styles and art direction

Read @aimikoda's prompt
Bottom Video
MiniMax H3 Prompt for Board Animation:

Use the provided character board @[board reference] as the sole visual source. Animate the existing static board itself while preserving the exact 16:9 composition, panel positions, character design, illustration style, color palette, and solid off-white background.

Keep the camera completely static for the entire video. No zoom, pan, tilt, reframing, parallax, or camera movement of any kind. The board itself must remain perfectly locked in place.

The large waist-up portrait on the left is the main animated focal point. Let the character naturally change pose within the existing portrait area: subtle head turns, gaze changes, shoulder and upper-body shifts, small hand or arm movements when visible, expression changes, blinking, breathing, and gentle hair and clothing motion. Avoid repeating a single idle loop. Make it feel like a short sequence of natural portrait poses.

The 3 full-body characters in the upper-right must remain in their exact panel positions and each perform a smooth full 360-degree turn in place, like a character turntable. Preserve their original proportions, outfit, identity, and panel boundaries.

The 3 environment panels in the lower-right must remain fixed inside their frames. Animate each scene subtly and differently, preserving its original setting and lighting. Let the character perform small natural actions appropriate to the environment, such as walking a few steps, shifting posture, sitting naturally, leaning, looking around, or reacting gently to the surroundings. Add restrained environmental motion such as rain, wind, moving clouds, reflections, foliage, drifting particles, or subtle lighting changes where appropriate.

Keep the bottom color palette completely static. Keep the solid off-white board background unchanged. Do not rearrange panels, expand scenes beyond their frames, add new elements, duplicate characters, redesign the character, or alter the layout. The result should feel like a polished animated artbook character board.

Top Video
MiniMax H3 Prompt for Character animation:

Use the provided character board @[board reference] as the sole visual reference for character identity, design, facial features, hair, proportions, outfit, colors, environments, and illustration style. Use it as reference only. Do not animate the board itself.

Create a short cinematic character introduction where the character comes to life through a sequence of polished animated shots inspired by the board’s content. Begin with the large waist-up portrait as a living shot, then move into the 3 full-body poses as brief character-focused moments, then transition into the 3 environment scenes as cinematic shots where the character naturally exists in those spaces. End on a clean composed hero shot of the character.

Let the character be the animated focus in every shot. Use natural motion such as blinking, breathing, gaze shifts, head turns, subtle pose changes, small gestures, gentle hair and clothing motion, walking, sitting, leaning, resting, or looking out depending on the scene. The 3 full-body moments should clearly show the character’s silhouette, outfit, and attitude. The 3 environment moments should show the character in different natural settings, different lighting conditions, and different natural poses that fit each location.

Use elegant cinematic camera movement, smooth transitions, and refined artbook-inspired presentation. Preserve strong character consistency and the board’s color language throughout. Minimal typography, atmospheric, polished, artistic, no redesign, no watermark.

Original post ↗

A tiny figure below a paper-cut palace

scene · MiniMax H3 per the author · Aug 21, 2026

A short prompt built on one clear ratio: a small figure, an endless staircase and a palace far larger than she is, all in the same paper material. Camera motion is limited to a slow push and orbit, with wind moving the paper, her hair and her clothes.

Giants, armies and epic scaleAnimated styles and art direction

The week's best new dramas and AI movies, in one email.

We are putting together a weekly email: the new dramas and AI movies worth your time, and any free-coin deals we find. Leave your address and you will get the first issue when it starts. No spam.