@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.
Drag the clip sideways
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.
@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.
@image1 is an absolute character lock: face, body, hair and the outfit are taken from it exactly, with no drift or reinterpretation in any shot. Filmed on a phone propped in place at a fixed angle, phone-vlog framing, with the pace carried by quick jump cuts between takes rather than by camera movement. Nobody ever reaches toward the camera. Photographic realism throughout: natural skin texture with visible pores and flyaway hair, true-to-life colour, soft even daylight, shallow natural depth of field. A fast-paced influencer-style "Get Ready With Me" video of someone preparing for a lunch hangout with their friends, focusing only on outfits and accessories. They turn to the camera, energetic, wearing something casual, and say, "GRWM for lunch with my friends—let's pick a cute outfit!" Quick jump cuts at a full-length mirror as they swap out of the casual outfit, with a stylish transition between the looks. They settle into the outfit from @image1 and add its finishing accessories, saying, "Okay this is giving effortless but cute!" They step into the shoes, settle the bag, and check themselves in the mirror with confident poses. They pick up their phone, smile at the camera, and say, "Alright I'm ready—see you at lunch!" Upbeat music, snappy edits, and strong influencer vlog energy throughout. Do not render: illustration, 3D render, cel shading, anime or cartoon styling, airbrushed or plastic skin, beauty-filter smoothing, on-screen text, subtitles, captions, watermarks or logos.
@image1 is MATEO — photocopy his exact face, wavy chestnut hair, stubble. IGNORE the tuxedo. Outfit: black mock-neck knit, charcoal trousers, silver watch. He MOVES. @image2 CLOSED foldable iPhone BACK: dual-camera island (two lenses + flash in a rounded square, NOT a 15 Pro diagonal bar), Apple mark. @image3 UNFOLDED INNER DISPLAY: book-style open, fold crease, tablet-like home screen. Same device. NEVER a MacBook. NEVER a candy-bar 15 Pro. 20-second 9:16 YouTube Shorts unboxing. Tech studio, clean desk, mic. Say “Mateo” only in the opening. [TIMELINE — 20 SECONDS] 0–3s: Box toward camera, peel tab, CLICK. His face: “Hey everyone, I’m Mateo — today we're unboxing the new foldable iPhone.” 3–8s: Overhead. Lid. He lifts CLOSED @image2, unfolds to @image3. “Wow… this is seriously clean.” 8–13s: Cable + docs. Inner screen on. “Inside: the foldable iPhone, USB-C cable, documentation.” 13–17s: He folds/unfolds, scrubs a clip on the inner display. “For creators, that inner canvas is the point — and it still pockets like a phone.” 17–20s: Folds halfway, to camera: “First impression? Premium build, beautiful inner display. Now I want a full editing test.” Smile. Audio: his lines, cardboard, hinge. No drowning music.
Your avatar runs a five-move dance medley on a retro TV soundstage. Music only, loud. See the video step for the full prompt.
@image1 (nine-look reference sheet) is an absolute appearance lock. Read it left to right, top row first: cells 1 through 9 are the nine looks, in that order. Face, body and skin tone are identical in every shot and taken from the sheet. Hair and clothing change only to the next cell's look, with no drift, reinterpretation or invented variation. [Style] Studio photography quality, warm beige solid-colour background, portrait lighting, smooth beat-based transformation rhythm. [Scene] The seamless warm beige studio background remains unchanged throughout. Front softbox portrait lighting, slight rim light on the hair edges. Background colour, light direction and exposure are identical from the first shot to the last. [Camera] Fixed vertical medium close-up from just above chest height, subject centred and facing the camera. The camera never moves. [Transition mechanism] Every change of look uses the same cut: at the end of a beat the subject turns their head to the side and whips their hair, the fast hair movement throwing motion blur across the frame and briefly covering the face, and the cut lands inside that blur. They turn back already wearing the next look. All nine transformations use this transition, with the spoken hairstyle name and a centred subtitle switching at the same moment. [Beats] All nine beats carry equal weight and equal screen time. None is rushed and none is held longer than the others. Beat 1, cell 1 — they face the camera and say "Long black straight." Turn and whip hair. Beat 2, cell 2 — the curls bounce with the turn as they say "Wool curls." Turn and whip hair. Beat 3, cell 3 — they raise a hand to say hello and say "Big waves." Turn and whip hair. Beat 4, cell 4 — the hair ends sway slightly as they say "Shoulder-length bob." Turn and whip hair. Beat 5, cell 5 — the ponytail swings with the movement as they say "High ponytail." Turn and whip hair. Beat 6, cell 6 — the hair falls gently as they say "Collarbone-length hair." Turn and whip hair. Beat 7, cell 7 — they say "Princess cut." Turn and whip hair. Beat 8, cell 8 — they make a heart shape to the camera and say "Egg roll curls." Turn and whip hair. Beat 9, cell 9 — they turn back with a wink and say "Super short hair." The subtitle pops up and they settle into a held final pose. [Audio] Upbeat beat-driven music underneath. Each turn and whip is paired with a whoosh of hair cutting through air and a click. The ending carries a rising freeze-frame sting. Spoken lines and subtitles are in English. Do not generate any text other than the nine subtitle lines named above. No logo, no watermark, no additional captions. The background never changes colour or texture. No camera movement, no zoom, no orbit, no push.
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.
IDENTITY LOCK — @image1: keep the exact face, bone structure, eyes, eyebrows, nose, mouth, skin tone, hairline, hairstyle, hair length, texture and colour, freckles or marks, apparent age, build and proportions identical in every frame. The attached @image1 outranks every word: where text and image disagree, follow the image. @image1 is an adult, the only person in the frame, and never turns into anyone else. The clothing comes from @image1 and never changes; fully dressed, ordinary closed shoes on at all times, never barefoot. THE ROUTINE — a set commercial hip-hop combination performed by a trained dancer, technical and hit-driven, not freestyle. GROOVE BASE: the bounce drives down on every count and releases on the and-count, never dropping. HITS: chest pop, shoulder snap and arm hit, each accelerating fast and stopping DEAD with a visible lock and micro-vibration. TUTTING: forearms build two or three precise right-angled frames in front of the chest and face, wrists breaking cleanly on the count. WAVE: one unbroken body wave fingertip-wrist-elbow-shoulder-chest-ribs-hips-knees, each segment a fraction after the one above. FOOTWORK: kick-ball-change into a cross-step behind, a heel-toe swivel, then a double-time shuffle under a still upper body. LEVEL CHANGE: a controlled drop into a deep squat on the beat, one hand sweeping a low arc, a push back up on the next count. TURN: a single clean inside turn on the ball of one foot spotting the lens, landing square on the beat back into the groove. DIRECTION — one continuous handheld-on-gimbal take: medium-wide in the key beam, snap-push on the first hit, orbit a quarter-circle through the tutting, glide low and lateral on the wave, drop to knee height on the footwork, sink and rise on the level change, crane up and back to a wide on the final hit; continuous screen direction, no cuts, no spinning camera, no teleporting. SCENE — a professional black-cyc dance studio: seamless dark floor with a soft sheen, black walls falling off into nothing, one hard key light from camera-left, a cool blue rim from behind, light haze, a faint floor reflection. Nobody else, no mirrors, no clutter. Let each beat play out fully; ONE continuous coherent take, no cuts. 0-2s ready bounce, head snaps up, first chest pop locks; 2-4s four hits each stopping dead, camera snap-push; 4-6s three tutting frames then a shoulder roll, camera orbits; 6-8s one full body wave, camera glides low; 8-10s kick-ball-change, cross-step, heel-toe swivel, double-time shuffle, camera at knee height; 10-12s deep squat level change with a floor sweep and controlled rise; 12-14s clean inside turn back into the groove plus two fast hits; 14-15s final hit locks hard, one arm into the beam, eyes to lens, camera cranes up and back. CHOREOGRAPHY — trained technique throughout: full arm extension, clean lines, dead stops on every hit, controlled weight, spotting on the turn, accurate foot placement, natural knee compression, realistic hair and fabric movement. ANATOMY — @image1 has exactly two arms and two hands with five fingers per hand, two legs and two feet, both feet contacting the floor cleanly on every landing. LOOK — ultra-photorealistic, 35mm and 28mm low, deep focus so the feet and the floor stay sharp, one hard key beam from camera-left, a cool blue rim from behind, visible haze, a faint floor reflection, deep true blacks, realistic grain, natural motion blur, cinematic shutter. AUDIO — music only, mixed loud and full so the music dominates the entire soundtrack from the first frame: an original hard hip-hop instrumental (heavy 808 sub, crisp snare on the backbeat, tight hi-hat rolls, a dark synth motif, stab accents that land exactly on the hits, a short filter build into the double-time bar). NO background or ambient sound of any kind, no studio room tone, no footsteps, no shoe squeak or scuff, no foley, no recognisable copyrighted melody, no spoken dialogue, no invented language. NEGATIVE — any background, ambient or foley sound, room tone, footsteps, shoe squeak or scuff audio competing with the music, music too quiet or not clearly audible; amateur or tentative movement, freestyle flailing, generic arm waving, swaying with no footwork, pausing to find the next move, moves blurring with no dead stops; the groove dropping; hits off the beat; the wave breaking into separate isolations; the turn wobbling or landing off the beat; the squat collapsing; feet hidden or cropped out of frame; floating, sliding or skating feet, gait glitching; the camera cutting, teleporting or spinning; a scene cutting off mid-action; jumping to a random unrelated scene; skipped or truncated beats; @image1 vanishing, dissolving, fading or teleporting; a second person or a duplicated @image1; @image1's face, hair, skin tone, body or age changing; a third hand, extra arm, more than two hands, extra or missing fingers; extra or missing legs, warped or morphing limbs, blurred or duplicated feet, body distortion; bare feet, barefoot, a single shoe, changing outfit; slow motion, speed ramps, repeated movement loops, aerial shot; plastic or beauty-filtered skin, anime, cartoon or CGI look; mirrors, studio clutter, on-screen text, captions, subtitles, numbers, readable signage, invented logos or watermark; only the elements explicitly described appear.