Made with OpenClips.AI

Your avatar fights a losing battle as they turn undead, ending as a frightening veined zombie. Music only, loud. See the video step for the full prompt.

Dark mist sweeps over your avatar and transforms their everyday look into a dramatic gothic horror costume, same face, confident and menacing. Music only, loud. See the video step for the full prompt.

A 22-second vertical 9:16 talking-product comedy ad. Photorealistic 3D character animation, cinematic. ASSET MAP — every reference is bound here and cited by tag below: @image1 — the magnesium product. This is the product's truth: its shape, material, colour, closure and every word printed on it. @image2 — the same product wearing its cartoon face and arms. This is the character design. PRODUCT LOCK — HIGHEST PRIORITY: the container is exactly the one in @image1. Preserve its shape and proportions, its material and colour, its cap or closure, and its label artwork exactly. Every word, number and symbol printed on it stays as photographed — same spelling, same typeface, same size, same position. Never add, remove, translate or re-typeset any lettering, brand name, logo, badge or seal. Wherever this text and the attached images disagree, follow the images. CHARACTER LOCK: the face and arms are exactly the ones in @image2 — same eyebrow weight and shape, same eye style, same mouth, same arm colour and thickness, same placement on the container. Do not redesign them. THE EYEBROWS ARE THE PERFORMANCE — two thick, solid, matte-black eyebrows shaped like bold brush strokes, clearly separated from the eyes and sitting well above them. They are the single most expressive element in the entire shot: they tilt, arch, furrow, raise and flatten on every emotional beat. They are never thin, never a mere eyelid line, and never absent from any frame. HANDS STAY CLEAR OF THE LABEL: the arms gesture beside the container and out to its sides, but never cross in front of its printed text. Every word on the product stays readable in every frame. THE BODY NEVER DEFORMS: the container does not stretch, squash, bend or hop. It stays standing on the counter. Only the face and the arms move. SET: a dark kitchen at three in the morning. The container stands on a stone counter. The only light is a cool under-cabinet strip and a faint glow from a doorway behind. A blurred kettle, a mug and a plant deep in the out-of-focus background. Quiet, still, nobody else in the room. CAMERA: one continuous shot, no cuts. It opens WIDE — the whole dark kitchen, the container small and alone on the counter — and pushes in slowly and steadily across the entire clip, arriving on a medium close-up of its face by the final line. The move is smooth and never stops. No handheld shake, no pan, no whip. VOICE: one voice, dry, weary and completely deadpan — something that has been waiting all day to say this. Not cartoonish, not sinister, not whispered. Ordinary speaking pace, with real pauses between sentences. PERFORMANCE AND LINES — it speaks these words, in this order, word for word, exactly once: 0–4s: wide shot. Arms crossed, eyebrows flat and low, already unimpressed before the clip starts. "Three in the morning. Calf cramp. And suddenly, you remember I exist." 4–9s: uncrosses its arms, one hand turning up in a small matter-of-fact shrug. Eyebrows level, plainly stating a fact. "I'm magnesium. I'm involved in hundreds of reactions in your body. Quietly. Every day." 9–14s: leaning into the grievance, eyebrows angling sharply down into a furrow, one hand pointing straight at camera on each question. Midway through this stretch a single tiny white spark flares off the cap for half a second and dies — it does not react to it. "That eye twitch last Tuesday? Restless legs at night? Your body was asking." 14–20s: one hand up flat in a stop gesture out to the side, then both hands opening outward. The eyebrows travel from firm and level to raised and almost pleading. "I'm not a painkiller. Don't wait for the cramp — take care of your magnesium first." 20–22s: the arms drop to its sides, the eyebrows settle flat, the eyes glance away, and it holds there as the camera settles on its face. "...I'll be here." AUDIO: quiet night room tone throughout, a faint fridge hum. One soft electrical tick on the spark. The voice. No music bed, no score under the dialogue. NEGATIVE: on-screen text, captions, subtitles, a watermark, a brand name, an invented logo, added or altered lettering, garbled label text, a hand covering the label, thin eyebrows, missing eyebrows, a human, hands belonging to a person, a second container, a second character, a face on any other object, photorealistic human eyes, teeth, lips, stretching, squashing, bending, hopping, the container leaving the counter, camera cuts, handheld shake, flickering light, horror atmosphere, sinister tone, spilled capsules.

3 steps

Your avatar turns to stone from the feet up, aware and struggling, ending as a weathered statue among the rocks. Music only, loud. See the video step for the full prompt.

Your avatar bursts into flames and rises as a fire-wielding villain, unharmed and menacing, wreathed in towering flames on a night city street. Music only, loud. See the video step for the full prompt.

@image1 fixes the object's identity: shape, proportions, construction, materials, hardware and colour are taken from it and must not drift, restyle or change at any point. The soiling seen in @image1 applies to the opening shot only. @image2 is the storyboard and is followed exactly — the same three shots in the same order, with the same framing, camera distance, hand positions and lighting. Style: photoreal macro cinematography, shallow depth of field with the focus plane riding the object, low-key directional light, cool neutral grade with warm specular highlights on the object's surfaces, fine natural grain, restrained handheld micro-movement throughout. Shot 1 — overhead close on the bench. Two hands in black nitrile gloves scrub the soiled object with a small detail brush through grey soapy foam, working along its length in short even strokes and lifting grime out of its seams. The camera eases in slowly from above. Shot 2 — cut to the same gloved hands holding the object up to camera, now fully cleaned. They turn it through a slow quarter rotation so the light travels across the cleaned surfaces. The camera stays where it is, focus riding the object as it turns. Shot 3 — cut to the cleaned object resting on the clean dark surface, hands out of frame. The camera pushes in slowly and settles, light raking across the object, holding to the end. Audio: close ASMR field recording — brush bristles working over the surface, wet foam shifting, small contact clicks against the bench, water and cloth, faint room tone. No music, no voice. Do not generate text, subtitles, captions, logos or watermarks. Do not substitute, redesign or restyle the object. Do not add tools, props, hands or people beyond those described. No cuts other than the two described.

3 steps150

ATTACHED IMAGES ARE MANDATORY AND FINAL. @image1 (poster establishing frame) is an absolute lock. Face, body, hair, build and skin are taken from it exactly and never drift, morph or get replaced in any beat. Its render style, palette, lighting, graphic system and framing are the look of the whole piece: photoreal premium editorial fashion photography, high-contrast clean studio light, visible skin texture and separated hair strands, real fabric weave and drape - never illustration, 3D render, cartoon or painterly. @image2 (wardrobe rail) is the sole source of truth for every outfit. Each beat below names one LOOK by label; dress the character in that exact set from the rail - its TOP PIECE, BOTTOM PIECE and FOOTWEAR - and nothing else. Never add, remove, substitute, restyle, re-cut or re-colour a garment, and never invent an outfit that is not on the rail. The character stays centred in frame for the entire piece. Not narrative - a beat-driven lookbook. The whole piece runs as one continuous take: outfits and layouts change on the beat through sharp in-frame glitch transitions, the display type tearing and shearing sideways with a red scan-line break and a brief datamosh smear, never a cut to a different place. Palette holds throughout: black and white base with red accents. Graphic system holds throughout, and its scale is deliberate. Only the oversized display headline carries readable letters - it is always huge, heavy and cropped by the frame edges. Every smaller editorial element is a clean geometric mark, not small text: solid bars, hairline rules, tick strips, barcode blocks, crosshair registration marks, corner brackets, thin frame borders and light modular grids, with red used only as an accent bar or rule. Keep these marks few, large and crisply edged. Do not render small paragraphs, fine print, dense numeral strings or any lettering too small to read cleanly. Beats, in order, each holding for the same length of time: 1. LOOK 01. Camera locked. They stand naturally, centred. The huge black type reading ONE HIT expands rapidly behind them and the editorial furniture snaps in on the beat, turning the clean studio into a bold magazine poster. 2. LOOK 01. Fast push-in to a close-up. They turn gently toward the lens, hair moving naturally, direct eye contact, calm and confident. The ground simplifies to black and white with film grain. 3. LOOK 02. The outfit changes on the beat. They hold a natural fashion pose, centred. The layout rebuilds completely around them into a new arrangement with an oversized headline, a barcode block and corner brackets. 4. LOOK 03. Medium shot. They take one confident step forward. The display type re-sets rapidly behind them; black and white stay dominant with stronger red accents. 5. LOOK 04. They make a gentle half-turn, hair swinging naturally. The ground flips to red and white - a huge headline, a scan-line band, heavy rules and a bold border. 6. LOOK 05. They lean slightly forward, relaxed and confident. Subtle low angle. Large red type slides into frame on the beat and locks hard against the frame edge. 7. LOOK 06. They drop into a crouching pose. Quick push-in. The layout completely rearranges around them, the headline and the marks landing in new positions. 8. LOOK 07. They stand again. Camera pulls back to a medium shot. The ground returns to black and white with a large headline and barcode elements. 9. LOOK 08. They walk confidently toward the lens. The ground switches to red and black; a new headline, scan-line bands and grid blocks stack in rapidly and the pacing tightens. 10. LOOK 09. Arms crossed, standing confidently, centred. Camera settles into a clean medium shot. The layout gradually simplifies until only a few premium magazine marks remain, a final graphic mark resolves, and the frame freezes on the last beat. Camera stays controlled throughout - locked holds, quick push-ins, subtle pull-backs, pose changes, graphic transitions and layout swaps. Maintain realistic hair movement, fabric physics and natural poses. Avoid: face drift, character replacement, body morphing, extra limbs, distorted hands, unstable proportions, messy typography, chaotic camera movement, random backgrounds, cartoon style, low-quality CGI, watermark, unwanted subtitles. No on-screen words other than ONE HIT.

3 steps534157.4K

Your avatar turns to ash and is carried off on the wind, aware and reacting with a dramatic raised hand, in a windswept field at golden hour. Music only, loud. See the video step for the full prompt.

MODEL: Seedance 2.5 — use this model specifically. Do NOT use any other model for this generation. FORMAT: 9:16 (vertical) QUALITY: High-fidelity, 4K render, sharp detail, no compression artifacts CHARACTER LOCK — HIGHEST PRIORITY: The character in every shot is exactly the person shown in @image1 — same face, facial features, hairstyle, skin tone, build and gender presentation throughout all 3 shots. No identity drift between shots. Do not beautify, slim, de-age or restyle them. Refer to them only as they/them in this prompt. @image1 is identity only — face, hair, skin — not the sunny café, not the cobblestone patio, not a seated still. Photocopy the person from @image1 into a rainy city bus-stop scene. ASSET MAP: @image1 — the character. Tell the following story in exactly 3 shots, same character throughout, with a clear visual beginning, twist and ending — never an ambiguous or faded-out close. STORY: The character stands alone at a city bus stop as rain suddenly starts pouring — everyone around them opens an umbrella, but they have none and huddle miserably under the tiny stop roof, getting splashed anyway. A stranger nearby silently steps closer and holds their umbrella out over them instead, sharing it without a word. They look up, surprised, then smile — the two end up walking off together down the street under the one shared umbrella, laughing, just as the rain eases and warm light spills from a café window ahead. STYLE: Warm, relatable live-action realism, soft naturalistic lighting, handheld documentary-style camera energy. Continuous real motion. Never freeze on the reference still. No on-screen text, no subtitles, no logos, no title card. AUDIO (native, synced, no music, no voiceover, no narration): rain pattering on pavement and the stop roof, city ambient noise, footsteps, the rustle of an umbrella opening, a warm shared laugh at the end. SHOT 1 (0-6s): Establish the character clearly at the bus stop in pouring rain. Handheld 35mm, medium-wide then push in. Rain starts hard in the first second — wet asphalt, droplets, people around them already sheltered under umbrellas. They have none. They huddle under the tiny stop roof, shoulders up, getting splashed anyway. The problem is immediately obvious. Crowd extras under umbrellas are background only. Face readable. SHOT 2 (6-12s): The twist. A stranger steps in from frame-left; their umbrella enters frame and tilts to cover them too (fabric rustle). Close enough to read their face: surprise, then a genuine warm smile. Rain still heavy. Handheld, eye-level. The stranger is a real second person sharing the umbrella — keep @image1's face locked on the character. SHOT 3 (12-18s): Resolution. Both walk together down the wet street under the one shared umbrella, relaxed and laughing (heard, not voiced as dialogue). Rain visibly eases. Warm golden light from a café window ahead. Handheld follow, then a clear last beat: they keep walking into that light — a finished ending, not a fade-out, not a still.

@image1 is LANA — photocopy their exact face and hair exactly as in @image1. IGNORE the clothes and accessories in @image1. Eyes visible. Lana MOVES. @image2 is the OUTFIT. Photocopy every garment, colour, fabric and accessory exactly as in @image2, worn in motion. Never add garments that are not in @image2. NEVER a studio cyclorama. 10-second 9:16. Ultra-real 35mm street fashion. Busy New York — yellow cabs, pedestrians, lights. LANA is ON the CROSSWALK (white stripes), walking toward camera mid-stride. The @image2 outfit moves with the stride. Crowd and traffic are BEATS. [TIMELINE — 10 SECONDS] 0–2s: Lana steps onto the crosswalk, the @image2 outfit moving, NYC behind. 2–6s: Walks toward camera on the stripes. Cabs. People around Lana. 6–10s: Closer, slight head turn, outfit still moving. Designed hold. Audio: city, footsteps, horns far. No lyrics.

A baby avatar performs a professional popping-and-locking routine in a studio. Music only, loud. See the video step for the full prompt.

Live-action photoreal footage in the grammar of a hand-held front-facing phone camera: real skin, real hair physics, natural indoor daylight, a constant small hand drift and breathing sway, shallow phone-lens depth. Not illustrated, not rendered, not stylised. @image1 is an absolute lock for the opening appearance — face, body and unstyled hair are taken from it exactly, with no drift or reinterpretation. @image2 is an absolute lock for the closing appearance — face, body and finished hair are taken from it exactly. @image1 and @image2 are the same person; the only thing that changes between them is the hair. The subject holds the phone at arm's length, looking straight down the lens with a flat, faintly unimpressed expression, mouthing along to the track, giving a small tilt of the head. Then their eyes flick off to one side and back, lips still moving with the words, shoulders loose, the unstyled lengths hanging heavily and barely moving with them. Then they lift one hand toward the lens and sweep it fast across the frame — the whole image smears into motion blur as the camera whips with it. Then the blur clears on the same person transformed, hair now finished and full, their head lifting into a bright open smile. Then they turn their head sharply one way and back so the styled lengths swing out wide and settle, fingers combing back through them. Then they push the hair back off one shoulder and lean slightly toward the lens, holding the smile while the lengths fall back into shape. On-screen text: the words new hair? in small lowercase serif, low and centred, present from the first frame to the last and unchanged across the transition. Audio: one upbeat pop track running unbroken from the first frame to the last — a bright four-on-the-floor kick, a bouncy synth-bass hook, handclaps on the offbeat and a bright plucked synth riff carrying the hook, confident and celebratory at a steady dance tempo. It plays quieter and slightly muffled across the opening, then opens up full and bright the instant the blur clears, and the whip lands square on a beat. Continuous hand-held motion throughout — nothing in frame is ever fully frozen. The single whip blur is the only transition; no other cut, wipe or dissolve. The subject mouths along to the track; no spoken dialogue, no voice-over, no singing voice and no lyrics — the track is instrumental. No added captions, titles, logos, watermarks or interface elements beyond the words above. No second person in frame. No cel shading, no illustration, no cartoon, no 3D render, no beauty filter, no skin smoothing.

4 steps

@image1, @image2 and @image3 are three wardrobe references of the same person. Face, body proportions, hair and identity are taken from them exactly and held without drift for the entire take. Each outfit is reproduced exactly as shown in its reference, down to footwear and accessories — no substitutions, no invented garments. Ultra-realistic throughout: real fabric, leather, suede, fur and skin behaviour, cinematic autumn colour grading, natural film grain, premium fashion-campaign realism. A fixed 90° overhead camera, locked off for the whole take. The subject lies on a luxurious autumn-toned Persian-inspired rug over walnut flooring, subtle upscale apartment furniture around the edges of frame, warm late-afternoon sunlight casting realistic shadows. A realistic wooden ceiling fan rotates continuously directly above them, hanging between the camera and the subject, its blades long and wide enough that each passing blade sweeps across their entire body and briefly occludes it. The fan is the only transition device. Wearing the outfit from @image1, they lie back on the rug, casually pick up a remote and turn their gaze up into the overhead lens. As a fan blade physically passes over their body, seamlessly reveal the outfit from @image2, the change landing under the blade itself, complete the instant the blade clears them, never a dissolve, never a fade, never a change in open frame. They settle into a new editorial pose, gaze still up into the lens. As the next blade physically passes over their body, seamlessly reveal the outfit from @image3, again changing under the blade and complete the instant it clears them. They adjust into another editorial pose, gaze still up into the lens. They hold that final pose, only the slow rise and fall of breathing and the faintest settling of fabric, while the fan keeps rotating overhead. Strictly no cuts, no camera movement, no zooms, no angle changes, no morphing, no identity drift, no outfit glitches, no warped hands, no duplicated limbs, no disappearing accessories, no text, no logos, no watermark.

4 steps1095.9K

Your avatar tours a haunted mansion at night as the house comes alive with dread, ending with a ghost appearing right behind them. Music only, loud. See the video step for the full prompt.

Match the ORIGINAL 15s 9:16 FILM shot-for-shot. Not a storyboard poster. Not a 2x2 grid. No on-screen campaign type. @image1 = THE BOTTLE. Photocopy its stopper, collar, liquid colour and every label word and mark exactly as printed. NEVER a can. @image2 = THE CRYSTAL ROCKS GLASS. Photocopy the glass exactly as shown — shape, cut and contents. Ignore bar bottles behind it. 15 seconds, 9:16, dark glossy studio, mirrored black floor, red-black bokeh. No people. AUDIO: ice rumble, spider-crack, explosive shatter, whoosh, pour, clink. No speech. 0.0–2.0s RECTANGULAR ICE BLOCK. The bottle stands fully trapped inside a tall clear ice prism, cracks already in the ice, shards at the base, reflection on the floor. Slow push-in. Product fills the block like the original can-in-ice. 2.0–3.5s SPIDER CRACK. Extreme close: ice in front of the bottle, radial fracture lines, a stream of water shooting across the crack. Macro. 3.5–5.0s TOP. Extreme close looking down on the bottle's stopper and collar, heavy condensation droplets, water splash above. No pull-tab. No can lid. 5.0–7.0s BURST. The bottle explodes out of the ice on a diagonal, flying toward camera, hundreds of ice cubes blasting radially, motion blur on shards, bottle sharp and readable in the center. 7.0–9.0s ORBIT. Bottle tumbling slowly in a vortex of ice and mist, wet glass, dark studio. Full 360 sense. Not standing still. 9.0–11.0s POUR. Macro: the spirit (its colour exactly as in @image1) pouring from the bottle neck into @image2's rocks glass packed with ice. Splash, bubbles, orange peel if it fits. High-speed. 11.0–15.0s HERO TWO-UP. Bottle LEFT, filled crystal glass RIGHT, ice chunks on a mirrored black floor, red-black bokeh, condensation, reflections of both. Slow push-in. Last second HOLD. No text, no slogan, no can. NEGATIVE: a can, pull tab, storyboard grid, posters, captions, extra bottles, people, warped bottle.

PRODUCT LOCK — CRITICAL: @image1 is the exact bottle. Photocopy its shape, cap, body colour, wordmark and every label line exactly as printed. Do not redesign. Do not turn it into another brand. One bottle only. Ignore any background or surroundings in @image1 — the film is a clean white studio. 10 seconds, 9:16, luxury beauty commercial. No people, no hands covering the bottle, no on-screen captions. AUDIO: soft water, liquid, lather, hair. No speech. No lyrics. 0.0–1.5s WATER REVEAL. Extreme macro, crystal-clear droplets falling slow-motion on soft white. Sparkle, forward push, shallow DOF. 1.5–3.0s HAIR. Close-up long silky hair flowing slow-motion, realistic strands, studio highlights. No face. 3.0–4.5s SHAMPOO TEXTURE. Creamy white shampoo flowing from THIS bottle (@image1), glossy ribbon, real viscosity. Bottle undistorted, label readable. 4.5–6.0s LATHER. Cream mixes with water into soft foam, tiny bubbles, droplets, macro. 6.0–7.5s SILKY HAIR. Camera glides through shiny hair swaying. No face. 7.5–10.0s HERO. The exact bottle from @image1 standing upright on glossy white. A few droplets, slow push-in. Last second: hold, sharp, centered, undistorted. NEGATIVE: another brand's bottle, distorted bottle, altered label, extra bottles, people, hands, watermark, subtitles, clutter, floating bottle.

Your avatar blooms into a lush red-flowering garden bush — head becoming one big red flower, aware and reacting, until no skin is left and the figure reads as a real leafy shrub of petals and green foliage blending into the hedges. Music only, loud. See the video step for the full prompt.

Smooth natural motion, clean commercial fashion-ad aesthetic, crisp detail, luxury catalogue presentation, soft studio lighting with realistic shadows under the figure and under the pieces. @image1 is the layout and the first frame, and it is an absolute lock. The entire layout stays exactly as shown: same background, same item panel, same captions, same crown icons, same wordmark, same framing, same position and same scale of every element. Face, hair, body, proportions, garments, materials and colours stay exactly as they are in that frame, with no drift or reinterpretation in any frame. Camera locked off on a tripod. No pan, tilt, zoom, tracking, dolly, handheld movement or reframing. The figure turns one full circle on the spot the way a fashion model turns on set, and the pieces inside the panel turn with them. Beats, each given equal weight: They begin in the pose of the first frame, then start to turn, stepping their feet around naturally and shifting their weight from one leg to the other, arms moving freely and shoulders relaxed. They continue round through profile, the turn unhurried and even. They come around to the back and glance over their shoulder toward the camera. They continue round through the other profile, still stepping the turn out on the spot. They return to face the camera and settle into a new, different confident pose, and hold it. Across the same turn, every piece inside the item panel rotates in place on its own vertical axis like a luxury product display render, revealing front, side and back. Each piece stays centred in its own slot, keeps its size, and completes exactly one full rotation, ending in its original orientation. The figure stays in the same spot and at the same size in frame throughout, feet staying on the floor with the contact shadow following them. Every garment continues around the side and the back, complete and unchanged, with each trim, fastening and fold consistent all the way round. Natural posture and realistic cloth physics: the garments, their hems, cuffs and edges, anything worn at the neck or the ankle, and the hair all swing and settle with the motion, then come to rest. The panel, its divider lines, the captions, the crown icons and the wordmark stay perfectly fixed and stay sharp and legible for the whole clip. Only the figure moves and the pieces rotate. No camera movement, no cuts, no identity drift, no face change, no garment substitution, no change to the panel layout or its type, no drifting or resizing of the panel, no sliding feet, no warped hands, no extra limbs, no disappearing accessories, no added scenery, no extra text, no logos, no watermark.

3 steps131.1K

integrated_multimodal_description: A photoreal fashion spec sheet that animates, held in one locked-off vertical frame. Warm off-white paper ground, soft even warm-neutral light, no cast shadows, everything in focus, no colour grading, no film grain. The frame never pans, tilts, pushes or zooms; every change happens inside a still frame. REFERENCES — the attached images outrank every word here. @image1 is an absolute character lock: face, body, hair and everything worn in it are taken from it exactly, with no drift or reinterpretation in any shot. @image2, @image3, @image4 and @image5 are the four wardrobe lineups; reproduce each one exactly as composed — the three views, their order and spacing, the stance, the annotation rules and labels, the bottom-left wording and its colour, and the palette chips. Nothing is redesigned, recoloured or re-laid-out. This prompt names garments only by slot and never describes them; the lineups are the only source of garment design, colour and material. Shot 1: The figure from @image1 stands full length, front on, centred, alone in the frame. They turn slowly and evenly on the spot, front to profile, feet pivoting under them. Small registration ticks blink on and off in the empty margins around them. Shot 2: The turn continues through the back view and returns to front on. As they settle, the measurement overlay draws itself on from the top of the frame downward: rules extend, end ticks snap in, leader lines run out to the margins, and the numeric callouts fade up in place. The unit caption settles last. Shot 3: Cross-dissolve. The lineup in @image2 resolves in place of the single figure, all three views arriving together. The side rules draw down the edges, the dimension brackets snap across the top, and the bottom-left wording and the palette chips fade up. Shot 4: Cross-dissolve to the lineup in @image3, the three views ghosting in over the outgoing ones before they clear. The same annotation build runs: side rules, then brackets, then the wording and chips, the wording arriving in its own colour. Shot 5: Cross-dissolve to the lineup in @image4. The same annotation build runs in the same order, the wording again in its own colour, the chips landing one after another from left to right. Shot 6: Cross-dissolve to the lineup in @image5. The same annotation build runs a final time, and the frame holds steady once the last chip has landed. Every movement runs forward in normal time: the subject always travels in the direction their body faces and nothing moves in reverse. Between the three views inside any one lineup, nothing about the person changes. Do not generate any text other than what the attached lineups already carry. No logo, no watermark, no subtitles, no captions, no brand names.

6 steps29826.9K

Your avatar melts like a candle, aware and reacting, until the whole body runs out into a pool of wax on the floor of their room. Music only, loud. See the video step for the full prompt.

CHARACTER LOCK — HIGHEST PRIORITY: the person is exactly the person in @image1. Preserve their facial identity, facial structure, eye shape and spacing, nose, mouth, exact skin tone, hairline, hairstyle, hair length, texture and colour, freckles and distinctive marks, apparent age, build and body proportions. Preserve their gender presentation exactly as shown. Photocopy their hair, eyes, skin, outfit and accessories exactly as in @image1. Do not beautify, slim, de-age or restyle them. Do not change the hair or the outfit. @image1 is identity and wardrobe — only their scale changes. No other people in frame. Use they / their / them only. SUBJECT LOCK — HIGHEST PRIORITY: Preserve the subject in @image1 exactly. If it is a person: their exact face, facial features, hairstyle, skin tone, build and gender presentation stay unchanged — only their scale changes. Do not beautify, restyle or alter identity in any way. @image1 is a REFERENCE still. Do not paste it as the opening frame. Do not open on a studio-grey headshot. CONTINUOUS MOTION REQUIRED: Real, visible physical motion for almost the entire 12 seconds — but the opening 2 seconds prioritize CLARITY over speed, so a first-time viewer immediately understands both the craft and the scale. SCENE: The subject, shrunk to a few centimeters tall, works at a tiny wooden bookbinding desk tucked into the gap between two real, full-sized books leaning against each other, their towering spines and pages framing the scene on both sides — the scale contrast unmistakable from the first frame. A tiny pot of glue, a miniature awl and a needle and thread sit on the desk. Warm desk-lamp light. 0-2s: ESTABLISHING SHOT — wide, clear view: the subject sits at the tiny desk stitching the pages of a small handmade book together with a needle and thread, the two giant books clearly visible looming on either side. Camera slowly pushes in, making both the craft (bookbinding) and the scale obvious immediately. 2-4s: They finish stitching the spine, set down the needle, and carefully brush glue along the book’s spine with a tiny brush, focused and precise. WOW MOMENT — TIME FREEZE + ORBIT (4-6.5s): They lift the finished tiny book and open its cover. As the pages begin to flutter open, everything suddenly FREEZES — the pages suspended mid-flutter catching warm light, tiny dust motes glittering around the subject. The camera performs a slow, cinematic 180-degree orbit around the frozen book and subject, revealing the two giant books towering on either side from multiple angles, shallow depth of field. This freeze holds for roughly 2 seconds. 6.5-7s: TIME RESUMES at full speed — the pages settle open, the tiny book now fully assembled in their hands. 7-9.5s: They close the book gently, walk it to the base of one giant book, and lean it upright directly against it — a small, complete, unmistakably-a-book object standing next to a real book, the scale payoff instantly clear. 9.5-12s: They step back, dust off their hands, and give a proud, satisfied smile toward camera, warm lamp light around them. SCALE REALISM: The giant books’ paper texture, cover material and page edges, and the tiny desk’s wood grain must all read at true scale relative to the tiny subject — every texture enormous next to them. Match lighting and color grade between the subject, the tiny book and the giant books throughout so the whole shot reads as one real photograph, not a composite. Frame wide enough that the establishing shot, the stitching, the freeze-orbit, and the final side-by-side book comparison all stay clearly visible. AUDIO: Generate synchronized natural audio for this scene — quiet thread pulling and stitching sounds, soft brush strokes, a gentle page flutter, near-silence during the freeze, then a soft whoosh as pages settle, and a satisfied contented sigh at the end. No music, no voiceover, no narration. No on-screen text, no watermark, no subtitles.

Your avatar transforms into the animal from the reference image in one dramatic cinematic take, close on the face. Music only, loud. See the video step for the full prompt.

3 steps

Two avatars perform a contemporary duet in a marble museum atrium. Music only, loud. See the video step for the full prompt.

Create an ultra-realistic lifestyle video as one seamless continuous-looking shot that loops perfectly. @image1, @image2 and @image3 are absolute locks. @image1 is LOOK 1, @image2 is LOOK 2, @image3 is LOOK 3. Face, hair, body, proportions, garments, materials and colours are taken from these references exactly, with no drift or reinterpretation in any frame. Every accessory shown in a reference stays present and unchanged whenever that look is on screen. Natural Instagram lifestyle Reel, relaxed and relatable. Soft autumn daylight, restrained warm grading, realistic skin, fabric and clothing movement. Not a luxury campaign, runway video or cinematic commercial. Camera on a locked tripod across the street, facing the subject straight-on and showing their full body. The framing never changes: no pan, tilt, zoom, tracking, dolly, handheld movement, reframing, stabilization drift or angle change. The subject stands naturally beside a pedestrian traffic-light pole on a simple city sidewalk, near the centre of frame, with the pedestrian signal clearly visible to their right. Keep the environment clean and minimal: a sidewalk, neutral urban background, road and one traffic-light pole. The pedestrian signal stays red for the entire clip. Beats, each given equal weight: They wait naturally beside the red signal in LOOK 1. A dark gray SUV passes through the extreme foreground and completely hides them. During full occlusion, instantly change to LOOK 2. LOOK 2 is revealed. They make one small, natural posture adjustment. A cream city taxi completely hides them. During full occlusion, instantly change to LOOK 3. LOOK 3 is revealed. They casually shift their weight. A black family sedan completely hides them. During full occlusion, instantly change back to LOOK 1. LOOK 1 is revealed again and they wait naturally beside the red signal, in the same stance, the same position and the same framing as the opening frame. There are exactly three outfit changes and three complete outfits. Every change is performed through a practical foreground vehicle wipe. Each vehicle passes close to the camera and completely hides their head, face, hair, hands, handbag, torso, legs and shoes. Switch outfits instantaneously only while they are 100% invisible behind the vehicle. Never show two outfits simultaneously. No morphing, dissolves, flashes or magical transformations. Their identity, hairstyle, proportions, standing position and background alignment remain perfectly consistent. All three vehicles travel one at a time in the same horizontal direction with natural speed, wheel movement, shadows and motion blur. The last frame must match the first frame so the clip loops with no visible seam: same outfit, same stance, same hand and bag position, same light, same red signal, same background alignment. Strictly no camera movement, visible cuts, outfit morphing, partial transformations, identity drift, face changes, sliding feet, warped hands, extra limbs, disappearing accessories, vehicles passing behind the subject, crowded scenery, a green signal, walking out of frame, text, logos or watermark.

4 steps099

Your avatar becomes a pale, fully translucent ghost, drifts off the floor, and fades from view at the end. Music only, loud. See the video step for the full prompt.

A two-step recipe for the talking-product format: an image step turns the uploaded packaging into a character, and a Seedance 2.5 step gives it a 25-second monologue in one continuous take with a moving camera. WHAT TO EDIT WHEN YOU RUN IT: the monologue in step 2. The words between the { } braces are an example — replace them with your own and keep the braces. Budget roughly 65-70 words for 25 seconds; more than that and the model compresses the closing lines until they are swallowed, which is the single commonest way this format fails. WHY THE SPEECH CLOCK IS IN THE PROMPT: without it the model runs the monologue too fast and leaves four or five seconds of dead air at the end, or crushes the last two sentences into two seconds. With the clock written out the pauses land within a few tenths of a second of where they are asked for. WHY THE HANDS ARE CONSTRAINED: on an early render of this format the direction said 'raises a single index finger' and the model raised the middle finger instead, and held it for ten seconds. A stubby cartoon hand cannot isolate one digit reliably, so both steps forbid pointing and finger-counting and use whole open palms. KNOWN WEAKNESS — READ BEFORE USING A REAL BRAND: step 1 does not always respect the printing on the packaging. On the example shown here it rotated the label ninety degrees and added two words that were not on the input photo. The prompt now forbids this explicitly, but check the step 1 output before spending on step 2, and re-run step 1 if the wording or its orientation has drifted. THE PROP IN STEP 2 IS OPTIONAL: the fizzing glass suits an effervescent tablet and nothing else. It is left in the prompt as a worked example of the technique, because a prop that visibly changes from the first frame to the last is a clock the viewer can read and it proves the take was never cut. Swap it for something that changes in your own product's world, or delete the block. PREVIEW: produced by this chain at 720p. Step 2 runs at 1080p for a finished ad.

2 steps

Upload an outfit photo and an avatar. Step 2 generates a clean outfit-only image; step 3 magically reveals your avatar wearing that exact outfit, keeping the same face. Music only, loud.

2 steps

Your avatar turns into fully transparent, see-through water and then suddenly collapses and splashes down into a clear puddle on a sunny poolside. Music only, loud. See the video step for the full prompt.

SHOOT 1 ELARA LOCK: @image1 is Elara — photocopy her exact face, hair and eyes exactly as in @image1, gold hoops, gold chain. Same person every frame. Never change their hair colour. Do not beautify. ROOM: the apartment in @image1, exactly as shown, at golden hour. Hallway beyond the door is the same home, warm wood, off-white walls. LOOK 1 (this whole take): exactly the outfit worn in @image1. SCENE: Elara films herself on her own front-facing camera while walking a warm dim hallway and through an open door into the sunlit living room, talking the whole time. Camera is her POV of herself, arm's length. Opens as she brings the camera up. One unbroken take. She crosses into late-afternoon sun, turns, stops in the light for the last line. 9:16 vertical. ~10 seconds. Real-time. No cuts. CAMERA: front-facing phone, swings up from below, holds ~50cm from face, soft bounce on steps, focus on face. ACTION: already walking. Delivers the line into the lens. One brief hair fix. Steps through the door into volumetric sun. Slows and stops for the last words, looking at the lens. VOICE, English, casual then quieter: "I don't know if I should say this, but the thing that scares me about 2026 is that nobody can tell I'm not real." AUDIO: NO MUSIC. Diegetic only — her voice close, footsteps on wood, fabric, room tone. No subtitles. No second person, no phone visible in frame, no logos, no UI. Living micro-motion every frame. Sun from camera-right in the room. SHOOT 2 ELARA LOCK: @image1 is Elara — photocopy face, hair and eyes exactly as in @image1, gold hoops. Same person every frame. Do not recast. ROOM: apartment in @image1, exactly as shown, late-afternoon sun from camera-right. Locked-off vertical UGC. No studio. @image3 = the sweater. Photocopy it exactly as in @image3 — knit, colour, collar, buttons and hem. @image5 = the shirt, exactly as in @image5. @image4 = the skirt, exactly as in @image4. Any label inside the back neck of the @image3 sweater stays exactly as it is; never invent a label. LOOK 1 (first half, still wearing): the outfit from @image1. LOOK 2 (after the cut): the @image5 shirt UNDER the @image3 sweater (collar and cuffs peeking) + the @image4 skirt + tan loafers. FORMAT: 9:16, 15 seconds, timed multishot, exactly ONE HARD CUT at 0:09. Instant cut. Same camera, room, light. Only wardrobe changes. She never speaks. CAMERA: fully locked off, low tripod, head-to-toe, one framing both sides. Phone autofocus hunts when the label comes close. 0:00–0:09 SHOWING. Already in motion. Holds the folded @image3 sweater, the @image5 shirt and the @image4 skirt. Lifts the sweater, lets it fall open, snap, show front and back. Drapes it, drops the skirt open so its details read, then the shirt. Steps in, folds the sweater inside back neck to the lens so its own label fills the frame. Focus hunts then locks on it. HARD CUT. 0:09–0:15 WEARING LOOK 2. Already dressed — no changing shown. The @image3 sweater over the @image5 shirt, the @image4 skirt, loafers. Tugs the hem, hand down the skirt, hair behind ear, weight on one hip, quarter turn, small closed-mouth smile. Mouth stays closed. No speech. AUDIO: NO MUSIC. Fabric rustle, loafers on parquet, room tone. No subtitles. No second person, no phone, no mirror, no extra logos. Fully clothed every frame. Living motion. SHOOT 3 ELARA LOCK: @image1 is Elara — photocopy face, hair and eyes exactly as in @image1, gold hoops. Same person every frame. Do not recast. ROOM: apartment in @image1, exactly as shown, late-afternoon sun from camera-right. Locked-off vertical. Same framing as the previous try-on. @image2 = the dress. Photocopy it exactly as in @image2 — fabric, colour, straps, neckline, detail and length. STARTING OUTFIT (first half): still wearing LOOK 2 — the @image3 sweater over the @image5 shirt, the @image4 skirt, tan loafers. AFTER THE CUT: LOOK 3 — only the @image2 dress. Tan loafers peek under the hem. FORMAT: 9:16, 15 seconds, exactly ONE HARD CUT at 0:08. Instant. Same camera and room. Only wardrobe changes. She never speaks. NO brand label close-up in this shoot — do not invent a brand label on the dress. CAMERA: locked off, low tripod, head-to-toe. 0:00–0:08 SHOWING THE DRESS. Already in LOOK 2, already lifting the @image2 dress. Lets it fall open full length, hem swaying. Small lift so its details read. Rotates to show the back, fabric swinging. No neck-label insert. HARD CUT. 0:08–0:15 WEARING THE DRESS. Already in the @image2 dress, no changing shown. Hand down the fabric, hair behind ear, hip pose, then one slow light pirouette — skirt and hair lift, not a trained dancer. Lands, closed-mouth smile. Mouth stays closed. No speech. AUDIO: NO MUSIC. Fabric swish, hem, loafers on parquet, room tone. No subtitles. No second person, no phone, no mirror, no logos. Fully clothed every frame. Living motion.

@image1 is an absolute lock for the object: shape, proportions, material, colour, surface pattern and markings are taken from it exactly, with no drift, restyling or reinterpretation in any shot. @image2 is the storyboard — follow it exactly, matching shot order, framing and camera distance panel for panel. Look: real handheld macro craft footage shot on a long lens. Natural workshop light, warm and directional. Shallow depth of field with a soft falloff, fine surface texture and real specular highlights on metal. Slight organic hand movement in every frame, live-action realism throughout. Hands hold the object over a worn wooden workbench, a piece missing from its upper edge, and a small steel tool taps once at the broken notch. Then the camera closes to macro as a fine tool bevels and cleans the raw edge, the shavings catching the light. Then the object is braced against a wooden block and a tool presses a soft lump of bright metal into the missing notch until it sits proud of the surface. Then a flat blade files the added metal flush, pale dust falling away across the bench. Then a burnishing tool sweeps along the seam and the metal brightens to a mirror sheen. Then the object turns slowly in soft even light, the polished metal repair carrying a travelling highlight along the mended edge. Camera: one move per beat and no more — a slow push in, a small reframe, a gentle tilt. Keep the object centred and continuously in frame. Audio: natural workshop sound only — tool contact, fine scraping, the settle of the object on wood. No music, no voice. Do not generate any text, captions, subtitles, titles, watermarks or logos. No faces and no bodies — only the working hands. No cuts to a different object, no second object, no duplicated copies of the object in frame.

3 steps

@image1 is an absolute lock for the object: shape, proportions, colour, material, surface finish and markings are taken from it exactly and must not drift, morph or be reinterpreted in any shot. @image2 is the storyboard and is followed exactly — same shot order, same framing, same hand positions, same camera distance, same bench and lighting. The object begins heavily soiled, dulled and discoloured, and is restored across the film to the exact condition shown in @image1. Nothing else about it changes. Five shots of equal length, hard cuts between them, no dissolves. Shot 1: gloved hands work a small abrasive pad back and forth across a soiled area in short firm strokes. Loose deposits lift, flake away and drift down out of frame. Shot 2: hard cut. The hands fold a dark cloth and draw it along the same area in slow overlapping passes, carrying away the loosened residue and leaving the surface beneath uneven and partly revealed. Shot 3: hard cut. A fingertip carrying a small dab of pale compound presses to the surface and works it in tight, even circles, leaving a soft opaque film in its track. Shot 4: hard cut. A clean cloth wrapped over the fingers draws steadily along the object and lifts the film away. The surface brightens behind the cloth as it travels, catching the lamp light in small moving glints along the cleaned edges. Shot 5: hard cut. The hands withdraw out of frame and the finished object is held alone while the lamp light travels slowly across its surface and settles. Camera: locked-off macro, one fixed framing per shot. No push in, no pull back, no pan, no tilt, no orbit, no handheld drift. Audio: close ASMR. The dry grain of the abrasive pad, the soft drag of cloth across the surface, the faint squeak of a fingertip working compound in circles, quiet room tone underneath. No music, no voice, no narration. Do not generate any text, captions, subtitles, watermarks or logos.

3 steps

CHARACTER LOCK — HIGHEST PRIORITY: The character in every shot is exactly the person shown in @image1 — same face, facial features, hairstyle, skin tone, build and gender presentation throughout all 3 shots. No identity drift between shots. Do not beautify, slim, de-age or restyle the face or hair. @image1 is identity only — face, hair, skin, necklace — not the sunny café, not the cobblestone patio, not a seated-at-a-table still, not a teacup, not a woven handbag. Photocopy the person from @image1 onto a city sidewalk as a street musician. ASSET MAP: @image1 — the character. Tell the following story in exactly 3 shots, same character throughout, with a clear visual beginning, twist and ending — never an ambiguous or faded-out close. STORY: A street musician sits on a sidewalk corner playing guitar to an almost empty street, their open case holding just a few scattered coins. People walk past without stopping, and the musician glances at the low afternoon sun, visibly discouraged, considering packing up. Then a small child breaks away from a nearby parent, walks up, drops a coin into the case, and starts spontaneously dancing to the rhythm — clumsy, joyful, completely unselfconscious. A couple of passersby stop to watch and smile. Word seems to spread in real time: within moments a proper small crowd has gathered around the musician, clapping along to the beat, more coins landing in the case, the musician now playing with visible renewed energy and joy, golden late-afternoon light warming the whole scene. STYLE: Warm, grounded live-action realism, soft golden-hour lighting, gentle handheld documentary-style camera energy, no stylization or slow-motion tricks. Vertical 9:16. Continuous real motion. Never freeze on the reference still. No on-screen text, no subtitles, no logos, no title card. AUDIO (native, synced, no added music track, no voiceover, no narration): live acoustic guitar playing throughout — the guitar itself carries the sound. Quiet street ambience at first, a coin dropping into the case, scattered footsteps slowing to a stop, then a swelling crowd murmur with rhythmic clapping by the end. SHOT 1 (0.0-9.0s): Establish the character clearly — sitting on the sidewalk corner playing guitar, case open with only a few coins, people walking past without stopping. Handheld 35mm, medium-wide then push in. The discouragement must be immediately obvious: a glance at the low afternoon sun, a tired shift in posture, maybe starting to reach for the case as if to pack up. Face readable. Passersby are background only. Guitar is being played for real — fingers moving on strings. SHOT 2 (9.0-18.0s): The twist — a small child approaches, drops a coin into the open case (heard), and begins dancing unselfconsciously to the music — clumsy, joyful. The musician’s surprised, delighted reaction is clear and readable on @image1’s face. One or two passersby slow down and stop to watch, smiling. Handheld, eye-level. Child, parent and passersby are real extra people — keep @image1’s face locked on the musician only. SHOT 3 (18.0-28.0s): The resolution — a small crowd has now gathered, clapping along in rhythm, more coins visibly landing in the case, the musician playing with clear renewed joy and energy, golden late-afternoon light across the whole scene. Handheld among the crowd then a clear last beat: musician still playing, crowd still there, a finished warm ending — not a fade-out, not a still.

CHARACTER LOCK — HIGHEST PRIORITY: the person is exactly the person in @image1. Preserve their facial identity, facial structure, eye shape and spacing, nose, mouth, exact skin tone, hairline, hairstyle, hair length, texture and colour, freckles and distinctive marks, apparent age, build and body proportions. Preserve their gender presentation exactly as shown. Photocopy their hair, eyes, any eyewear and outfit exactly as in @image1. Do not beautify, slim, de-age or restyle them. Keep glasses only if @image1 shows them; never add any. @image1 is identity and wardrobe — only their scale changes. No other people in frame. Use they / their / them only. SUBJECT LOCK — HIGHEST PRIORITY: Preserve the subject in @image1 exactly. If it is a person: their exact face, facial features, hairstyle, skin tone, build and gender presentation stay unchanged — only their scale changes. Do not beautify, restyle or alter identity in any way. @image1 is a REFERENCE still. Do not paste it as the opening frame. Do not open on a studio-grey headshot. CONTINUOUS MOTION REQUIRED: Real, visible physical motion for almost the entire 12 seconds — but the opening 2 seconds prioritize CLARITY over speed, so a first-time viewer immediately understands the setting before the action builds. NO MUG. There is no coffee mug, no cup, no steam anywhere in the shot. The scale anchor is a stack of real, full-sized hardcover books. SCENE: The subject, shrunk to a few centimeters tall, works at a tiny handcrafted workshop bench built from scrap wood and matchsticks, laid out with a small anvil and a row of miniature tools — a tiny hammer, wrench and screwdriver — under a small glowing desk lamp. The bench sits on the corner of a real, full-sized desk, right beside a short stack of normal-sized hardcover books (worn cloth and paper spines, a few millimetres of page-edge showing) that tower over the tiny subject and anchor the scale contrast. Warm lamp light from one side. 0-2s: ESTABLISHING SHOT — wide, clear view of the full miniature workshop bench and its tools, the subject standing at it holding a small piece of metal up to the lamp light, inspecting it. The giant stack of hardcover books is visible in the same frame beside the bench, making the scale contrast unmistakable from the very first seconds. Camera slowly pushes in. 2-4s: The subject sets the metal piece onto the anvil, picks up the tiny hammer, and begins striking it in a steady rhythm — each hit sends a spark into the air, the metal slowly glowing and taking shape, sweat visible on their brow. WOW MOMENT — TIME FREEZE + ORBIT (4-6.5s): On the final, hardest strike, a bright shower of sparks bursts upward. At the instant the sparks reach their peak height, everything suddenly FREEZES. Dozens of glowing sparks hang suspended in mid-air around the subject, who is frozen mid-swing with the hammer raised. The camera performs a slow, cinematic 180-degree orbit around the frozen subject and sparks, revealing the full workbench layout and the giant book stack from multiple angles, shallow depth of field. This freeze holds for roughly 2 seconds. 6.5-7s: TIME RESUMES at full speed — sparks shower down and fade out, the hammer completes its strike with a final ringing clang. 7-9.5s: The subject sets down the hammer, picks up the now-finished tiny piece — a small gleaming gear or medallion — with miniature tongs, and holds it up to inspect it against the lamp light. 9.5-12s: They walk it over and place the finished piece down directly beside the giant book stack, creating a clear side-by-side size comparison against the book spines and page edges, then step back, wipe their brow, and give a satisfied nod and thumbs-up toward camera. SCALE REALISM: The workbench grain, the miniature tools and the books’ cloth spines, paper edges and cover boards must all read at true scale relative to the tiny subject — every texture enormous next to them. Match lighting and color grade between the subject, the sparks and the desk throughout so the whole shot reads as one real photograph, not a composite. Frame wide enough that the establishing shot, the hammering, the freeze-orbit, and the final side-by-side reveal all stay clearly visible. AUDIO: Generate synchronized natural audio for this scene — ambient workshop quiet, rhythmic metallic hammer clangs, a sharp sizzle with each spark, near-silence during the freeze, then a whoosh as sparks shower down, and a satisfied exhale at the end. No music, no voiceover, no narration. No on-screen text, no watermark, no subtitles.

@image1 is the absolute character lock for the opening state: face, body, hair and clothing are taken from it exactly, with no drift or reinterpretation. @image2 is the absolute lock for the state after the transition: hair and makeup match it exactly. The person, the clothing and the location are the same before and after — only the hair and makeup change. One continuous handheld shot at arm's length, the subject framed from the chest up. They face the lens with the chin dipped slightly and lift one open palm toward the camera. The palm rises and sweeps across until it fills the frame completely in motion blur. As it clears to the opposite side they are revealed in the state of @image2, already settled and facing the lens. They raise one hand and rake the fingers slowly back through the hair from the temple, releasing the lengths to fall. They turn the head to one side and back, the hair swinging out and resettling against the shoulder. They lower the hand and square up to the lens, a small closed-lip smile arriving as the last of the hair comes to rest. Constant handheld micro-movement throughout, the framing breathing gently as if the camera is held in one hand. The palm crossing the lens is the only transition — everything on either side of it runs as one unbroken take. The camera holds its distance: no push in, no pull back, no orbit. No spoken dialogue, no on-screen text, no watermark.

2 steps80

Your avatar runs a five-move dance medley on a retro TV soundstage. Music only, loud. See the video step for the full prompt.