Alex Neyef retweeted
Bye bye Seedance 2.5, my new favourite model is WAN 3.0, available in @makeugc
1. It's like 2,5 times cheaper
2. It has better quality
3. It does super realistic stuff
4. It bends video right to the prompt
Made this cool horror story with just 1 try lol
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Who wants this coffee maker?
Unlimited Seedance 2.5 on @makeugc
prompt
THE MORNING COFFEE FIX
PRODUCT A sleek, compact, portable espresso coffee maker in matte cream-white with champagne-gold accents, a transparent espresso cup, a circular power button, and a removable water chamber
A sleek, compact, portable espresso coffee maker in matte cream-white with champagne-gold accents, a transparent espresso cup, a circular power button, and a removable water chamber.
The product must maintain identical proportions, color, materials, button placement, and physical design throughout the entire video.
MAIN CHARACTER grok-image-b2d8ad30-25bf-4268-b73d-88b73b6bebaf
Her personality is cheerful, slightly sleepy, playful, and naturally expressive.
CHARACTER CONSISTENCY: Maintain her exact face, hairstyle, outfit, body proportions, and voice throughout every shot.
ENVIRONMENT
A cozy modern apartment kitchen during a bright, peaceful morning.
White marble countertops, cream cabinets, warm wooden shelves, a small potted plant, a ceramic coffee mug, and sunlight streaming through large windows.
The kitchen feels genuinely lived-in rather than artificially staged.
CINEMATOGRAPHY
Shot on a high-quality smartphone with natural handheld movement.
Authentic UGC influencer aesthetic with occasional cinematic product close-ups.
Natural autofocus adjustments, realistic exposure changes, subtle camera shake, soft morning sunlight, and accurate skin textures.
No artificial beauty filters, excessive color grading, or CGI appearance.
SHOT 1 — THE MORNING STRUGGLE
00:00–00:03
Medium handheld shot.
The woman stands in her kitchen, looking slightly sleepy. She runs one hand through her messy hair and looks toward the camera.
She gives a small, tired laugh.
DIALOGUE (sleepy, conversational):
"Okay... I seriously need coffee."
PERFORMANCE: A natural half-second pause after "Okay." She smiles slightly while speaking.
AUDIO: Quiet kitchen ambience, distant birds, soft fabric movement.
SHOT 2 — PRODUCT INTRODUCTION
00:03–00:06
She reaches across the countertop and picks up the portable espresso maker.
She holds it casually toward the camera, turning it slightly to show its design.
DIALOGUE (playful, slightly amused):
"But honestly... this little thing?"
[Brief pause. She glances at the product.]
"It's been saving my mornings."
Her expression is genuine and relaxed.
SHOT 3 — ADDING COFFEE
00:06–00:09
Close-up of her hands opening the coffee chamber.
She adds a measured portion of ground coffee and carefully secures the compartment.
The camera captures realistic textures of the coffee grounds.
DIALOGUE (casual, speaking while working):
"Just add your coffee..."
[Short natural pause.]
"Super easy."
SFX: Soft coffee container clicks, gentle countertop sounds.
SHOT 4 — ADDING WATER
00:09–00:12
Macro close-up of warm water being poured into the espresso maker's water chamber.
Tiny reflections and natural liquid movements are visible.
She carefully closes the chamber.
DIALOGUE (relaxed):
"A little hot water..."
[Brief pause.]
"And that's pretty much it."
SFX: Natural water pouring, subtle lid click.
SHOT 5 — PRESSING THE BUTTON
00:12–00:15
Close-up of her thumb pressing the circular power button.
The espresso maker activates.
She watches it with a curious little smile.
DIALOGUE (slightly excited):
"Okay, watch this."
[One-second pause.]
SFX: Realistic activation click and quiet brewing mechanism.
SHOT 6 — THE COFFEE REVEAL
00:15–00:18
Beautiful macro shot.
Rich, dark espresso slowly pours into the transparent cup.
A golden crema forms naturally on the surface.
Soft morning sunlight catches the flowing coffee.
DIALOGUE (genuinely impressed, slightly quieter):
"Ohhh... look at that."
[She pauses to admire the coffee.]
SFX: Gentle espresso extraction, liquid pouring.
SHOT 7 — FIRST SIP
00:18–00:21
Medium close-up.
She lifts the small espresso cup and takes a natural sip.
Her eyes close briefly.
She lowers the cup and gives a satisfied smile.
DIALOGUE (soft, happy):
"Mmm..."
[One-second pause.]
"That's actually so good."
Her reaction should feel spontaneous, not rehearsed.
SHOT 8 — THE PERSONAL TOUCH
00:21–00:24
Handheld medium shot.
She leans casually against the kitchen counter while holding her espresso.
She looks directly into the camera.
DIALOGUE (friendly, conversational):
"And I don't even have to leave my kitchen."
She laughs softly at the end.
PERFORMANCE: Natural facial movement, relaxed shoulders, authentic eye contact.
SHOT 9 — PRODUCT CLOSE-UP
00:24–00:27
The espresso maker sits on the marble countertop beside the freshly brewed espresso.
The camera slowly moves closer, revealing the matte cream finish and champagne-gold accents.
The woman reaches into the frame and gently picks up her coffee.
DIALOGUE (off-camera, playful):
"Yeah... I'm keeping this."
A quiet, genuine chuckle follows.
SHOT 10 — NATURAL ENDING
00:27–00:30
Medium shot.
She smiles at the camera, holding her coffee in one hand and the portable espresso maker in the other.
She gives a tiny approving nod.
DIALOGUE (warm, lightly teasing):
"Morning officially saved."
She takes another small sip and smiles.
The camera remains handheld until the final frame.
DIALOGUE AND VOICE DIRECTION
All dialogue must sound like a real person casually filming a morning vlog.
Natural American English accent.
Warm, youthful female voice.
Realistic breathing between sentences.
Slight hesitation before certain words.
Authentic pauses lasting approximately 0.3–1 second.
Natural variations in speaking speed and pitch.
Occasional soft laughter and spontaneous reactions.
Perfect lip synchronization.
Realistic facial expressions matching the dialogue.
No robotic speech, exaggerated enthusiasm, or commercial announcer voice.
IMPORTANT: The woman must not speak continuously. Allow moments of silence while she handles the product, watches the coffee brew, and takes a sip.
AUDIO DESIGN
Natural kitchen ambience, birds outside the window, coffee grounds pouring, water pouring, product button clicks, espresso brewing, gentle sipping, subtle breathing, and soft laughter.
No background music.
No voice-over narrator.
All dialogue comes directly from the woman on screen.
REALISM REQUIREMENTS
Photorealistic human anatomy, realistic skin pores, natural hair movement, physically accurate hand interactions, believable espresso extraction, natural coffee crema, authentic liquid physics, and consistent product design.
The character must never appear computer-generated.
All actions must follow real-world physical logic.
NEGATIVE PROMPT: Robotic dialogue, unnatural pauses, exaggerated acting, fake smiles, porcelain skin, plastic faces, CGI humans, inconsistent character identity, distorted hands, extra fingers, floating objects, changing product shape, unrealistic coffee physics, artificial liquid effects, excessive cinematic transitions, subtitles, text overlays, watermarks, background music. Style: old iPhone camera quality, low-resolution texture, rough digital noise, compression artifacts, imperfect exposure, natural shadows, no cinematic lighting, no film look, no beauty filter, no CGI appearance.
slightly imperfect framing, realistic photo. Natural skin texture visible with pores, tiny sweat sheen.Subtle natural skin texture, pores, fine details and realistic complexion. CAMERA:
DV 16mm tape camcorder handheld aesthetic. A gorgeous glamorous European fitness influencer films herself directly by hand in selfie-cam and first-person style. Keep realistic hand shake, slightly crooked framing, delayed autofocus, awkward zoom-ins and zoom-outs, occasional motion blur, and small framing mistakes where part of her face briefly slips out of frame. When switching from handheld footage to a fixed external angle, use a clean jump cut as if the camera had already been positioned before the shot began. The camera itself must never appear on screen.
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Bro, this is NOT how Halloween was supposed to go
Unlimited Seedance 2.5 on @makeugc
prompt
Video Prompt: STYLE: Photoreal raw flash footage, like a found Halloween night tape. Shot on a cheap early-2000s consumer camcorder with a fisheye lens: strong barrel distortion, harsh on-camera flash/light blasting whatever is close to the lens with blown-out highlights and hard shadows, everything past a few meters falls off into an underexposed, near-black background.
Heavy camcorder noise, video grain, slight interlacing, smeared motion blur, auto-exposure pumping, occasional focus hunting, slightly crushed and green-tinted digital color. Chaotic documentary realism: the operator is a person running and reacting in real time, frantic handheld whip-pans, crash zooms, the camera bumping into things, subjects shoved right up into the fisheye lens. Meme energy: absurd moments played totally real.
FAST pace: hard jump cuts every 0.5-1 second, no transitions. No on-screen text, no timestamp overlay, no titles, no logos, no watermark. CAST: @image (15) the devil woman.
Dale the werewolf: a huge, hulking, shaggy grey-brown werewolf with glowing yellow eyes, a long snarling snout full of teeth, and matted fur. He wears a torn red varsity-style bowling jacket hanging off his massive frame, shredded at the sleeves. The skeletons: a trio of full-body human skeletons with bleached off-white bones, wearing mirrored aviator sunglasses and black tactical vests. One of them wears a bright orange baseball cap and white sneakers. The elderly witch: a frail, wrinkled old woman with wild grey hair, wearing a black feathered cape and sharp cat-eye glasses, with a pointed witch's hat and a crooked grin. The Headless Horseman: a tall rider in a burgundy velvet coat with no head, seated on a powerful black horse with a flowing mane. Shot 1 (0-2.2s): Flash-lit fisheye shot at a gas station at night, the devil woman bored against an old pickup, camera close on her face; out of the black behind her, the werewolf sprints into the flash and leaps straight at the lens; she swings up a shotgun and fires, muzzle flash whiting out the frame, camera knocked sideways. Shot 2 (2.2-2.8s): Whip-pan into a dark alley: three skeletons on a low wall lit only by the flash, one raises his rifle and they jump down at the camera. Shot 3 (2.8-3.6s): The devil woman leaps out of a black archway right past the lens, her red tulle skirt and sheer sleeves flaring and filling the frame, firing a rifle. Shot 4 (3.6-4s): Flash-blown close-up of muzzle flash and sparks, camera shaking. Shot 5 (4-5s): Running handheld through a dark farmyard toward a shed with propane tanks; the elderly witch zooms past the lens on a ride-on lawnmower, cape flapping, straight into the tanks.
Shot 6 (5-6s): Massive fireball overexposes the camera; the witch is blasted up into the black sky; operator falls back. Shot 7 (6-6.6s): Pitch black cornfield, flash pops on: her broom stabs into the dirt right in front of the lens. Shot 8 (6.6-7.8s):
City street at night, camera on the sidewalk; the Headless Horseman gallops inches past the fisheye lens holding a flaming jack-o'-lantern, fire flaring the image. Shot 9 (7.8-8.8s):
A parked old sedan explodes, camera recoils, image clips to white then auto-exposes back. Shot 10 (8.8-12.2s): Packed barn Halloween party, flash on, crowd in witch hats and bedsheet ghosts screaming and holding up phones.
Dance battle filmed from the front row: the devil woman, in her full devil costume with horns and heels, hits a sharp freeze right in the fisheye lens; the skeleton in the orange cap and white sneakers does a headspin, then collapses into a pile of loose bones.
Crowd goes insane, camera shaking. Shot 11 (12.2-14s): Camera drops to the floor: harsh flash on scattered bones, the skull rolls straight into the lens and stops, distorted by the fisheye, background black. Hold. AUDIO: Raw camcorder mic audio, clipping and distorted: wind, werewolf roar, shotgun blast, people yelling 'YO!', lawnmower engine, explosions blowing out the mic, hooves, party crowd screaming, bones clattering, skull rolling. No music, no score.
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
She microwaved it for exactly 2 minutes
Made with Unlimited Seedance on @makeugc
Prompt👇
Native 23-second vertical 9:16 smartphone video, one continuous handheld take with no cuts, real-time speed. Shot by a friend standing in the kitchen doorway, phone at chest height, about 2.5 meters from the subject. Real everyday footage: natural daylight, real physics, believable human reactions, slight phone shake, auto-exposure, mild sensor noise, no color grading, no filters. No text, no captions, no logos, no readable labels. The story plays strictly in chronological order; every action is caused by the previous one.
SUBJECT
Mia, a clearly adult woman, 24. Warm light-tan skin with real texture, long dark brown hair in a high messy ponytail, small gold hoop earrings, light natural makeup, small beauty mark under her left eye. Fitted cropped white ribbed tank top, loose light grey sweatpants, white socks. Naturally sexy in an effortless, believable way, full, large, natural-looking bust, realistic body proportions. Expressive but natural face, dry sense of humor.
LOCATION (fixed for all 23 seconds)
Small bright modern kitchen. White cabinets, light grey speckled countertop along the back wall, light wooden floor. A stainless steel microwave sits on the countertop in the LEFT third of the frame and never moves. A window with white blinds on the far RIGHT gives soft daylight from the right. On the counter in the right background, within her reach: a white kettle, a bowl of green apples, and one unopened plain white popcorn bag with no text. The camera position never changes. Nobody else is visible.
POPCORN PHYSICS
Real fluffy white popcorn with a pale butter tint, every piece about 2 cm, dry and light. It pours like a dry waterfall, bounces, rolls and spreads across the floor, then piles up and rises gradually and evenly around her. It never passes through her body, the counter or the cabinets. No smoke, no steam, no fire.
TIMELINE
0.0-2.0s HOOK. Mia stands in the center of the frame, a step to the right of the microwave, half-turned toward it. The microwave is running, its inside light glowing orange; the glass door is completely packed with popcorn pressed against it from inside, and the door visibly bulges and rattles with rapid loud popping. Mia takes one quick step back, both hands raised halfway in front of her chest, palms open, eyes wide, mouth slightly open. Real startled reaction.
2.0-5.0s THE BURST. With a sharp plastic snap the latch gives and the door swings open to the left. A thick continuous stream of popcorn bursts out, spills over the countertop edge and pours onto the floor like a waterfall, scattering toward her feet. Mia flinches, turns her face slightly away for a moment, then looks back, frozen in disbelief, hands still raised.
5.0-12.0s THE FLOOD. Popcorn keeps pouring out of the microwave without stopping, far more than could fit inside. It spreads across the whole floor and rises steadily: ankles at 6s, knees at 9s, mid-thigh at 12s. Mia slowly backs up one step until her lower back touches the counter behind her on the right, lowers her hands and rests her arms at her sides on the rising popcorn. Her face goes from shock to blank resignation; one slow blink. The camera wobbles slightly as the friend holding the phone laughs quietly.
12.0-15.0s THE STILLNESS. The level rises from mid-thigh to her waist, then the stream stops; the last few pieces tumble out of the open microwave and roll onto the pile. Mia stands still, arms resting on top of the popcorn, looks down at it, then back at the camera with a long blank stare.
15.0-19.0s PUNCHLINE ONE. She picks up a small handful of popcorn with her right hand, eats a few pieces and chews calmly. She looks straight into the camera and says in a flat, dry, deadpan voice: "Okay. Two minutes was too LONG." The microwave beeps three times. She keeps chewing, one eyebrow slightly raised.
19.0-23.0s PUNCHLINE TWO. She turns her upper body to the right, reaches behind her with her right hand, picks up the unopened plain white popcorn bag from the counter, turns back and holds it up next to her face toward the camera. With a tiny mischievous smile she says: "Round TWO?" The friend behind the camera immediately shouts "NO!" while laughing, and the phone shakes. Mia keeps holding the bag up, smiling, as the video ends.
PERFORMANCE AND MOTION
Natural real-time movement with weight and inertia: eyes react first, then head, then shoulders and body. Visible breathing, small asymmetric gestures, relaxed fingers, five fingers on each hand; the bag is held firmly in her right hand by its top edge. Moderate mouth movement when she speaks, only her upper teeth show. Head turns stay small, her face is always visible and stays identical, hair never covers her face. Her feet stay planted when she is not stepping; she never jumps, slips or falls.
CONSISTENCY
One continuous shot: same camera position, same kitchen, microwave always on the LEFT, window always on the RIGHT, same daylight, same outfit, hair and earrings from the first to the last frame. Only one popcorn bag exists, on the counter until she picks it up. The microwave glass and the window show no reflection of her or the camera.
SOUND
Phone microphone in a small kitchen: microwave hum, rapid popcorn popping, door rattle, sharp latch snap, dry rushing hiss of popcorn pouring and pattering on the floor, her short startled gasp, a friend's quiet laugh, soft trickle of the last pieces, quiet crunching, three microwave beeps, her two lines, the friend's laughing shout "NO!". No music.
STYLE
Feels like a real viral home video that happened by accident. Natural, believable, funny. Not polished influencer content.
🤝 Paid partnership · 🤖 Made with AI
why is nobody selling this
Made with Unlimited Seedance on @makeugc
prompt here:
Create a native 15-second, horizontal 16:9 realistic street video in the style of casual handheld footage. No reference image. No music. No on-screen text.
CORE RULES
One main character, Hana, a clearly adult Japanese office worker, 24 years old. One secondary adult: a man in his 40s, office worker, appears in shots 2 and 4. One location: a quiet residential side street in a Japanese suburb during a HEAVY DOWNPOUR on a weekday morning. The rain is the second main character: dense, clearly visible, loud, pouring the whole video. The central object is her transparent full-body dome raincoat: a rigid, clear vinyl bell-shaped dome, like a giant clear dome umbrella with no handle, covering her from the top of her head to just below her knees, with a thin clear rim at the bottom. It has exactly two round side openings at elbow height that always stay empty; her arms and hands stay INSIDE the dome for the whole video. The dome keeps the same shape, size and transparency in every frame. The rain hits only the outside; inside she is completely dry, with no fog or condensation. Nothing is written anywhere.
CHARACTER
Hana, 24, slim with realistic body proportions and a full, large, natural-looking bust, naturally sexy in an effortless, believable way. Shoulder-length straight black hair tucked behind both ears, light natural makeup, soft pink lips. Fitted dark navy blazer buttoned once, white blouse with a slightly open collar, knee-length dark navy pencil skirt, sheer skin-tone tights, black low-heeled pumps, a plain black backpack on both shoulders, a thin silver watch on her LEFT wrist. Calm, quietly pleased, like someone who solved the rain problem.
FACE
Her face stays clearly readable through the clear vinyl at all times, even in the heaviest rain. Same face and features in every shot. Mouth closed or a small closed-lip smile; no teeth showing, also while eating. Natural blinking every 2 to 4 seconds. Eyes look where she goes, at her food or phone, never into the lens.
VOICE
No speech. Ambient sound only.
CAMERA:
Authentic casual footage captured with an older consumer digital camera. Handheld DV 16mm camcorder aesthetic. An unseen person operates the camera from a plausible nearby position. Preserve the viewpoint, framing and shot sequence described in the scene. Camera reactions feel human and slightly imperfect. The recording device stays outside the frame.
CAMERA HAND
The operator stands 2 to 4 meters away under cover of their own, camera at chest height. The operator, their hands, umbrella, shadow and reflection never appear. No raindrops on the lens.
HANDHELD:
Natural hand shake, slightly crooked framing and small framing corrections. Brief motion blur during camera movement, occasional delayed autofocus after reframing and subtle vertical movement from the operator's body. Keep faces and important actions readable. Avoid perfectly stabilized or gimbal-like movement.
LOOK:
Soft digital-tape look with subtle vintage DV character. Soft fine detail, faint digital noise, mild compression, slightly muted colours, restrained contrast and imperfect white balance. Preserve natural skin tones, pores, wrinkles, fabric texture and realistic surface reflections. No beauty filters, plastic skin, glossy commercial finish or heavy colour grading.
LIGHTING:
Natural overcast daylight from a pale cloud-covered sky. Soft shadows, gentle light falloff, ordinary skin tones and realistic surface reflections. Keep faces and surroundings clearly exposed while preserving natural differences in brightness. No sunset tint, dramatic rim lighting or artificial glow.
The sky is a heavy dark grey, slightly darker than an ordinary cloudy day, so the rain streaks stand out.
WEATHER
Heavy downpour, clearly visible in every shot. Dense thin diagonal rain streaks at many depths: close to the camera, around Hana and far down the street, most visible against the dark door, bushes, house walls and the dark sky gaps. The far end of the street is softened by a grey haze of falling rain. The asphalt is covered with a thin moving layer of water; every puddle is constantly broken by dense ripples and small bouncing splashes. Water runs in a small stream along the curb and drips in streams from roof edges and the entrance overhang. Rain drums on the dome: drops splash on its top, merge into running rivulets down the curved sides and pour off the bottom rim in a thin ring of falling drops. The rain is real 3D rain in the scene, never a flat uniform overlay over the image, and it never hides faces. Rain stays outside the dome and outside the covered entrance.
VISUAL MATCH
The image must look exactly like this: an ordinary person filming a neighbour during a real downpour on a grey Japanese morning. Full-body or knee-up framing, subject in the centre, camera at chest height, slightly tilted horizon. Muted grey, wet asphalt, dull green and navy palette; the brightest element is the glossy clear dome streaming with water. Mild noise, slight softness, no graphics.
SETTING
A narrow wet asphalt street with a white painted edge line. On one side: a beige two-storey apartment building with a covered entrance, grey metal door with a frosted glass panel and a pale tiled step of two stairs. Along the street: a low dark green metal mesh fence, tall trimmed green hedges behind it, wooden utility poles with sagging wires, two-storey houses fading into rain haze. No signs, no road markings with text.
BACKGROUND
Three planes in every shot. Front: Hana in the dome. Middle: flooded asphalt with rippling puddles and the green fence. Far: poles, wires, houses, rain haze. Anchors stay identical: the building on the LEFT, the green fence and hedge on the RIGHT. The background never warps.
MOTION
Calm, unhurried movement. Eyes lead, then head, then shoulders. Natural breathing. Short steps in the pencil skirt, knees following the toes, slight hip sway. The dome moves only with her body, rigid, with a small sway on each step, and its lower rim never touches her legs. She never hurries: she is the only calm person in the downpour.
DEFECT PREVENTION
Hands: shot 1 – both hands INSIDE the dome hold the inner lower rim, then rest inside at waist level. Shot 3 – both hands hold the rice ball inside the dome at chest height. Shot 4 – RIGHT hand holds the phone inside the dome at chest height, LEFT hand on the backpack strap. Hands and arms never pass through the vinyl or the side openings. Five fingers per hand; fingers never merge with each other, objects or her body; the camera hand is never visible.
Face: when the dome passes over her head in shot 1 her face stays visible through the clear top. Rivulets on the dome run down the sides, never as a sheet over her face. During bites the rice ball stays below her nose and her lips close while chewing. Her face, teeth and features stay identical in every shot; head turns under 45 degrees.
Body and legs: she steps down the two tiled stairs RIGHT foot first, LEFT follows. While walking, weight transfers heel to toe. In shot 3 she stands still with weight on both feet. Exactly two arms and two legs; joints bend only naturally; feet never slide on the wet ground.
Objects: one dome, one backpack worn under the dome, one watch on her LEFT wrist, one rice ball (plain white rice wrapped in plain dark nori, no wrapper) in shot 3 only, already in her hands at the start of shot 3, one black smartphone with a dark unreadable screen in shot 4 only. The man has one clear umbrella in his RIGHT hand and one black briefcase in his LEFT hand. Objects never vanish, duplicate, change shape or colour, or pass through bodies. No readable text on any object.
Continuity across cuts: same outfit, hair tucked behind ears, backpack, watch, the same rain intensity in every shot; building on the LEFT, fence on the RIGHT. Each shot starts from a pose that logically follows the previous one.
Physics and environment: no water, drops or fog inside the dome. Her hair, blazer and skirt stay perfectly dry. The man is getting soaked: dark wet patches on his shoulders and trouser legs, water streaming off his umbrella edge. No reflections of her or the camera in glass or windows. Nobody else appears. The background never warps or shifts.
STORYBOARD
SHOT 1 (00:00.0–00:03.5) Operator on the street facing the entrance; Hana full-body on the tiled step under the dry overhang, door behind her on the LEFT; heavy rain pours just beyond the overhang edge in the foreground.
00:00.0 START POSE (cover frame): both arms raised inside the dome, hands on the inner rim above her shoulders, the dome already over her head. Weight on both feet. Small pleased smile.
00:00.0–00:00.7: she pulls the dome down; eyes look up at the dome top.
00:00.7–00:01.3: the rim slides past her shoulders and backpack and settles below her knees with a soft rigid bounce.
00:01.3–00:02.2: hands rest inside at waist level, small shoulder roll, she looks out at the pouring rain. FACE: confident small smile.
00:02.2–00:03.5: she steps down the two stairs RIGHT foot first, LEFT follows; the instant she leaves the overhang the rain drums on the dome top and rivulets start running down.
SHOT 2 (00:03.5–00:06.0) Operator ahead of her, slowly walking backwards; Hana full-body walking toward camera on the flooded street, fence on the RIGHT. Far behind, the man with the clear umbrella walks away, hunched against the rain.
00:03.5–00:04.6: two short steps through the thin water layer, small splashes at her pumps; rain streams down the dome.
00:04.6–00:05.3: she tilts her head up to watch water pour down the vinyl. FACE: amused small smile.
00:05.3–00:06.0: she looks ahead and keeps walking calmly.
SHOT 3 (00:06.0–00:10.0) Operator 2 meters away, knee-up framing; she stands by the green fence, hedge behind, rain bouncing off the fence rail.
00:06.0 START POSE: standing still, weight on both feet, rice ball in both hands at chest height, eyes on it.
00:06.0–00:07.2: small bite, lips close. FACE: chewing slowly, eyes half closed with pleasure.
00:07.2–00:08.4: she glances up at the water sheeting down the dome, small content nod.
00:08.4–00:10.0: second small bite, chewing; rain pours off the dome rim around her knees.
SHOT 4 (00:10.0–00:15.0) Operator in front, full-body; the man walks on the LEFT half of the frame, Hana on the RIGHT half, side by side toward camera. The man: grey suit, white shirt, plain white face mask, short black hair, clear umbrella in RIGHT hand, black briefcase in LEFT hand, shoulders soaked.
00:10.0–00:11.5: they walk; water streams off his umbrella edge onto his shoulder. Hana scrolls her phone, LEFT hand on the strap.
00:11.5–00:13.0: he walks hunched, she walks upright and relaxed. FACE: calm.
00:13.0–00:14.2: she glances at his wet shoulder, then back to her phone. FACE: small smug closed-lip smile.
00:14.2–00:15.0: they keep walking toward camera through the downpour; she stays perfectly dry.
SOUND
Loud steady downpour, rain drumming hard on the vinyl dome with a hollow patter, rain hammering the man's umbrella, water gurgling along the curb, splashing footsteps, faint nori crunch in shot 3. No music, no voices.
HANDS AND FEET
Relaxed hands with natural finger spacing. Pumps stay on her feet with flat contact on every step. The man's steps match his walking pace.
CONTINUITY AND REALISM
Everyday imperfections: water spots and small scratches on the vinyl, a crease in the skirt, puddle ripples, rain haze. The dome looks like a real manufactured product, not a CGI bubble.
FINAL FEEL
A real neighbour casually filming a woman in a weirdly genius raincoat during a heavy downpour. Believable, slightly funny, everyday. Not polished influencer content.
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Seedance 2.5 vs WAN 3.0. Same prompt. Made on @makeugc
Both landed the triple axel. Physics is solid on both.
But look at the skin. WAN went full plastic, everything too smooth and glossy.
Seedance looks like real competition footage.
Which one would you use?
To be continued...
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
⚠️People urgently need to come up with a raincoat for cycling.
Made with Seedance 2.5 on @makeugc
Prompt
Create a 15-second photorealistic video, 16:9, using the attached image as the opening frame and strict visual reference.
Preserve the elderly man's exact identity: swept-back silver hair, round glasses, gray mustache and goatee, charcoal suit, blue shirt, navy tie and black backpack. Preserve the silver bicycle and its large transparent rain capsule, including its rounded shape, metal supports and clear panels.
Preserve the white-and-brown Jack Russell terrier and its miniature transparent rain capsule. The dog's capsule encloses its head and torso, with FOUR separate openings for its legs. Each leg stays through its own opening throughout the video. The dog runs on its own paws; its capsule has no bicycle, wheels or motor.
SETTING:
A rainy Tokyo business district with modern glass office buildings, warmly illuminated interiors, green sidewalk railings, pedestrian crossings and commuters carrying umbrellas. Wet asphalt reflects windows and traffic lights. Steady natural rainfall, gray daylight, realistic subdued colors. The scene feels like an unusual invention casually filmed on a real city street.
0–5 SECONDS:
Begin with the exact composition of the reference image, immediately in motion. The elderly man pedals smoothly toward the left at a relaxed cycling pace. The dog runs alongside him on the camera-facing side, slightly ahead of the front wheel, maintaining a clear lateral gap.
The camera tracks parallel to them in a medium-wide side view. Keep the entire bicycle, roof of the capsule and all of the dog's paws inside the frame. Buildings and pedestrians move naturally across the background.
5–10 SECONDS:
Continue the uninterrupted tracking shot. The man keeps both hands on the handlebars and looks calmly ahead. His legs drive the pedals continuously; the crank, chain and wheels move consistently with his speed.
The terrier maintains an energetic, natural running gait matching the bicycle's pace. Its paws make clear contact with the road, producing small splashes. Its lightweight capsule gently bobs with its torso while retaining its exact shape. The four leg openings remain fixed in the shell, allowing free leg movement.
Raindrops collect and slide down both transparent capsules. Preserve realistic visibility and subtle refraction through the plastic.
10–15 SECONDS:
The camera gradually moves slightly ahead and turns toward the subjects, transitioning smoothly from side profile into a front three-quarter tracking view while continuing to move with them.
Reveal more of the elderly man's face through the curved clear windshield. Keep the dog fully visible beside him, separated from the bicycle wheels. Both continue forward at the same steady pace.
Finish with both subjects still moving naturally through the rainy street.
CAMERA AND STYLE:
One continuous take. Natural street-video realism, subtle handheld micro-movement, believable perspective and restrained motion blur. No dramatic zooms, slow motion, cuts or exaggerated cinematic effects. The absurd rain capsules are treated as ordinary everyday objects.
AUDIO:
Natural rain pattering against clear plastic, tires rolling over wet asphalt, light bicycle mechanism sounds, quick soft paw splashes and distant city traffic. No dialogue, narration or music.
CONSISTENCY:
Keep the man's face, age, clothing and backpack unchanged. Keep the dog's breed, markings and proportions unchanged. Maintain the exact construction of both capsules. No morphing, extra limbs, paws passing through solid plastic, sliding feet, floating wheels or distorted bicycle geometry. No collisions or pedestrians obstructing the main subjects. No captions, logos or watermarks.
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Same hair
Different energy
Made by WAN 3.0 on @makeugc
Ur dream comes to reality
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Ghostface has better footwork than half the Premier League and I have the footage.
15 seconds. Five horror icons. One ball. Zero defending.
He picks it up near the halfway line and just goes. Jason throws his whole body at him and misses. Michael Myers gets turned so badly he needs a minute. Freddy commits to a slide tackle with the claws and stays on the grass for the rest of the clip. Pennywise never even gets close.
The finish is cold. Low, bottom corner, keeper nowhere.
The part people won't like: none of this was filmed. No costumes, no pitch, no stunt team. It's an AI video, and the robe movement, the shadows, the ball spin all hold up frame by frame.
Not long ago these models couldn't keep a ball attached to a foot for two seconds. Now they handle a full dribbling run with five characters, contact and a slide tackle.
Halloween content used to mean renting costumes and begging friends to show up at a park on a Saturday.
Now it's one prompt and a coffee.
So be honest. Is this the most fun Halloween has looked in years, or the moment you stopped trusting anything you see online?
Harry just got schooled by Malfoid 💀
Made with Wan 3.0 on @makeugc
Workflow in the comments 👇
#aivideo #MalfoidArt
🤖 Made with AI
Made with Wan 3.0 here: makeugc.ai/?ref=alexneyef
Workflow:
1. Found a clip of soccer tricks and used it as the video reference.
2. Created the Malfoid reference in ChatGPT. Described Harry in words because his image kept getting blocked.
3. Described the new locations in the prompt.
4. Did the whole video edit in Wan 3.0.
Prompt:
Edit the supplied 12-second Video 1. Replace the two people and the environments while preserving the original action, camera work and lighting.
CHARACTERS
Replace the woman wearing the pink football shirt with the woman from the supplied reference image. Match her face, physique, very long platinum-blonde hair and clothing: a flowing black robe with emerald-green lining, white collared shirt, fitted grey waistcoat, green-and-silver striped tie, short grey pleated skirt, black tights and black shoes.
Replace the man wearing black with a young male wizard with messy dark hair, thin round black glasses and a small lightning-shaped forehead scar. Dress him in a burgundy-red textured athletic top over a charcoal undershirt, black leather protective chest armour with crisscross front lacing through metal eyelets, padded dark shoulders and subtle gold trim. Complete the outfit with dark trousers and practical black lace-up shoes.
Use the supplied image as an appearance reference for the woman only. Create the male character from the written description above; no image reference is provided for him. All poses, gestures, expressions, gaze directions and movements must follow the corresponding people in the source video. Keep both identities and outfits consistent across every shot, including profile views and partial occlusions. Hair and clothing move naturally with the original performance. Their hands remain available for the original actions; do not add handheld accessories.
ENVIRONMENTS
Replace the existing locations in their original order:
The opening grassy field becomes the grounds of an ancient British Gothic castle beside a large dark lake. A recognizable cluster of tall stone towers, steep slate roofs and pointed turrets rises in the distance.
The football court beneath the urban bridge becomes a broad stone courtyard beside a monumental medieval viaduct, with high arches, weathered masonry and castle towers behind it.
The next green football field becomes a large grass sporting pitch surrounded by tall wooden spectator towers with colourful banners. Three elevated golden ring-shaped goals stand at each distant end.
The final paved court becomes an open castle courtyard enclosed by Gothic cloisters, pointed stone arches, columns and weathered flagstones.
Change environments only at the original cuts. Match the source perspective, horizon height, ground orientation and playing-space dimensions. Keep structures outside the players’ paths. Preserve believable foot contact, ball contact and foreground occlusion.
PRESERVE THE SOURCE PERFORMANCE
The source video is the sole motion and timing reference. Preserve every run, dribble, kick, feint, jump, turn, reaction and interaction frame for frame.
Keep the original football unchanged, including its colour, size, rotation, trajectory, bounces and exact contact timing.
Preserve the original duration, shot order, cuts, playback speed, framing, camera movement, handheld shake, lens perspective and motion blur. Do not introduce new actions, shots, slow motion or transitions.
LIGHTING AND APPEARANCE
Match each source shot’s original light direction, softness, exposure, white balance, contrast and shadows. All replacement environments remain under the same overcast daylight as the source footage. Adapt both characters and their clothing to that light.
The result should resemble ordinary live-action footage of real people playing football in these locations. Natural facial anatomy, realistic skin, believable fabrics and consistent contact shadows. Match the source sharpness, grain and compression. Avoid doll-like faces, artificial skin smoothing, CGI rendering, magical effects and dramatic colour grading.
Alex Neyef retweeted
Forget ChatGPT and Nano Banana for character references
I made this yoga video with Seedance 2.5 on @makeugc. This time I DID use a reference — and it looks just as realistic as my videos made without one.
No sketch. No image generator. The reference came from a 4-second Seedance video.
Here’s the workflow:
1. Describe your character directly in Seedance 2.5.
2. Set the duration to just 4 seconds to keep the credit cost down.
3. Choose 720p or 1080p.
4. Generate 3–4 short clips, the same way you’d generate a few images. Adjust the description until you like the result.
5. Take a clear screenshot, or cut out your character and remove the background.
6. Upload that image as your reference for the next video.
You’re creating the character in the same model you’ll use to film them. For me, that kept the natural look I wanted instead of the glossy AI look I kept getting.
Want the prompt for this yoga video?
PROMPT 👇
Character = MILA — a beautiful American Instagram-style brunette model and performer from reference 1, reference 2 and reference 3.
Scene: Quiet terrace at sunrise, simple railing, potted plants, open sky, soft golden light. Peaceful atmosphere.
Style: RAW HANDHELD REALISM, casual camera sway, micro-shakes, frequent reframing, natural autofocus, occasional off-center framing, slight motion blur, realistic colors, candid unposed feel.
00:00–00:04: She kneels and rolls out a yoga mat as the sun rises.
00:04–00:08: Sits cross-legged, eyes closed, hands on knees, breathing calmly.
00:08–00:12: Performs a slow sun salutation, arms raised, then folds forward toward her feet.
00:12–00:16: Steps back into plank, holding steadily with controlled breathing.
00:16–00:21: Transitions into downward dog, gently stretching her heels toward the mat.
00:21–00:25: Moves into warrior pose, arms extended, gaze focused.
00:25–00:30: Holds the pose, relaxes her arms, and gives the camera a peaceful smile before the natural cut.
Audio: Only natural ambience—birds, breeze, distant rooster or traffic, mat movement, and soft breathing. No music. 16:9, photorealistic, natural movement, consistent identity and outfit.
#AIVideo
🤖 Made with AI
Made with Seedance 2.5 on @makeugc using the same prompt (native 30-second video, no editing)
PROMPT ⬇️
Character = MILA — a beautiful American Instagram-style brunette model and performer, 20. Long dark-brown hair in a high ponytail, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
She wears a cobalt-blue supportive sports bra with thin straps and a deep scoop neckline showing cleavage, paired with matching ultra-short fitted athletic shorts. Bare shoulders, arms and midriff. White ankle socks, sneakers off. She speaks with an American accent.
Scene: Quiet terrace at sunrise, simple railing, potted plants, open sky, soft golden light. Peaceful atmosphere.
Style: RAW HANDHELD REALISM, casual camera sway, micro-shakes, frequent reframing, natural autofocus, occasional off-center framing, slight motion blur, realistic colors, candid unposed feel.
00:00–00:04: She kneels and rolls out a yoga mat as the sun rises.
00:04–00:08: Sits cross-legged, eyes closed, hands on knees, breathing calmly.
00:08–00:12: Performs a slow sun salutation, arms raised, then folds forward toward her feet.
00:12–00:16: Steps back into plank, holding steadily with controlled breathing.
00:16–00:21: Transitions into downward dog, gently stretching her heels toward the mat.
00:21–00:25: Moves into warrior pose, arms extended, gaze focused.
00:25–00:30: Holds the pose, relaxes her arms, and gives the camera a peaceful smile before the natural cut.
Audio: Only natural ambience—birds, breeze, distant rooster or traffic, mat movement, and soft breathing. No music.
16:9, photorealistic, natural movement, consistent identity and outfit.
#AIVideo
Alex Neyef retweeted
🍬Does candy make kids hyper?
Made with unlimited seedance on @makeugc
🤝 Paid partnership · 🤖 Made with AI
Alex Neyef retweeted
Good morning, Folks. Happy Saturday.
Ok, last night must been epic! And I have no idea where I managed to get an inflatable pink flamingo on a Friday night 🤷♀️ C'est la vie😂
Made with Seedance 2.5 on @makeugc
🤝 Paid partnership
Alex Neyef retweeted
⚠️This is what games will look like when AI takes over
Does anyone know any top Asian games? I really like the visual.
Seedance 2.5 on @makeugc
PROMPT:
Create a 30-second photorealistic stealth-game video in 16:9.
REFERENCE PRIORITY
is the authoritative reference for the adult female protagonist: preserve her face, hairstyle, physique, outfit, footwear and backpack throughout.
-Image2 is the storyboard for the first half.
-Image3 is the storyboard for the second half.
Read each storyboard from left to right, top to bottom. Follow all six panels of Image2, then all six panels of Image3.
The storyboard images define actions, locations, camera compositions and game interface. If the protagonist’s appearance differs between references, Image1 takes priority.
IMPORTANT
Produce full-screen moving scenes. The storyboard grids are planning references only: never display panels, borders, collages or split screens. Animate the actions between the illustrated moments naturally rather than presenting a slideshow.
VISUAL STYLE
Realistic third-person stealth gameplay with cinematic camera work, believable human motion and natural daylight. Modern university athletics administration wing: sage-green walls, pale terrazzo floors, oak doors with glass panels, black-framed windows and indoor plants.
Maintain consistent room geography: the office entrance connects to the corridor, the sage-green drawer cabinet stands against the side wall, and the solid-front oak desk stands near the windows. Preserve these positions across camera changes.
The protagonist wears her backpack on BOTH shoulders throughout. Both straps remain attached and the backpack moves naturally with her body.
SHOT SEQUENCE
0–3 seconds:
Start immediately with the protagonist concealed beside a corridor corner. The camera looks past her shoulder toward a male staff member walking away. She cautiously peeks around the corner and waits until he disappears from view.
HUD status: “STAY HIDDEN”, then “WAIT FOR PATROL”.
3–6 seconds:
She approaches the office door and looks through its glass panel. Show her checking the empty office before touching the handle. She listens briefly, then reaches for the handle.
HUD status: “CHECK THE OFFICE”.
6–9 seconds:
Close view of her hand pressing the handle, followed by a medium shot as she opens the door and slips inside. The camera follows smoothly through the doorway. She leaves the door slightly ajar.
HUD status: “OPEN DOOR”, then “ENTER QUIETLY”.
9–12 seconds:
She moves to the sage-green cabinet, crouches beside it and pulls out a drawer. Show the drawer sliding along its rails and her hands searching inside.
HUD status: “SEARCH DRAWER”.
12–15 seconds:
She spots her pink-cased smartphone, grasps it and lifts it from the drawer. The phone is clearly visible. Her expression briefly shows relief. She quietly pushes the drawer closed.
HUD status: “PHONE RECOVERED”.
Change the main objective to “ESCAPE UNSEEN”.
15–18 seconds:
Footsteps approach. She freezes, turns toward the doorway and lowers herself behind the cabinet before a female lecturer enters. The lecturer has a short dark bob, a teal blazer and cream trousers, matching Image3. She carries folders and walks toward the bookcase without noticing the protagonist.
HUD warning: “TEACHER RETURNING”.
The ALERT meter rises to amber because of the nearby threat, without reaching full detection.
18–21 seconds:
While the lecturer faces away, the protagonist moves into concealment behind the solid-front oak desk. Use a clear wide shot to establish that the desk blocks the lecturer’s line of sight. She holds the recovered phone close and watches the lecturer carefully.
HUD status: “STAY OUT OF SIGHT”.
21–25 seconds:
The lecturer remains turned toward the bookcase. The protagonist quietly moves from behind the desk toward the office exit, staying outside the lecturer’s field of view. The camera tracks her movement smoothly. She crosses the doorway into the corridor.
HUD status: “MOVE TO THE EXIT”.
25–28 seconds:
From the corridor, she gently pulls the office door closed. The latch clicks softly. She glances back, then relaxes as the ALERT meter falls to zero.
HUD status: “CLOSE THE DOOR”.
28–30 seconds:
Medium frontal tracking shot as she walks down the corridor. She raises the recovered phone at chest height and gives a small relieved smile. Keep both backpack straps visible.
Main objective changes to “MISSION COMPLETE”.
Final status: “PHONE RECOVERED”.
GAME INTERFACE
Match the HUD design shown in Image2 and Image3: small translucent charcoal panels, thin pale outlines, cyan accents and pink or amber alert indicators.
Keep the interface anchored to the screen, with identical placement and typography across every shot:
- Top left: “HEALTH”, “STAMINA”, “ALERT”.
- Top center: the current mission objective.
- Bottom left: a circular mini-map showing the protagonist’s position and orientation.
- Bottom right: “C CROUCH”, “E INTERACT”, “SHIFT SPRINT”.
- Lower center: the current action or warning.
Initial objective: “RECOVER YOUR PHONE”.
After collecting the phone: “ESCAPE UNSEEN”.
Final objective: “MISSION COMPLETE”.
HEALTH stays constant. STAMINA changes subtly with movement. ALERT responds to the lecturer’s proximity and returns to zero after escape.
Keep all English text sharp, correctly spelled and stable. Never cover the protagonist’s face or hands with interface elements.
AUDIO
Quiet corridor ambience, distant footsteps, soft sneaker steps, backpack movement, door-handle click, drawer sliding and a restrained suspense soundtrack.
A subtle notification sound when the phone is recovered, a brief warning tone when the lecturer approaches, and a short success chime at the end.
No dialogue or voice-over.
CONTINUITY
Natural movement, realistic weight and hand contact with objects. No teleportation, duplicate characters, extra limbs or changing outfits. The phone remains in her hand after collection. Doors and drawers move physically and consistently.
The lecturer never directly sees the protagonist.
Maintain a playful, tense stealth-game atmosphere and a complete beginning, middle and ending.
🤝 Paid partnership · 🤖 Made with AI
This is roughly what my dreams look like when I’m at peace.
Seedance 2.5 on @makeugc
PROMPT:
Create a 10-second vertical 9:16 video closely matching the original reference video's atmosphere, camera movement, pacing and surreal alpine scenery. Use the provided image as the exact opening frame and preserve the new woman's appearance, lavender jacket, dark cargo trousers and quad roller skates.
ONE CONTINUOUS REAR TRACKING SHOT:
An adult woman roller-skates downhill along a winding mountain road. The camera follows smoothly from behind at approximately waist height, maintaining a consistent distance. Keep her full body and both skates visible throughout, with space beneath the wheels. She stays in the lower central portion of the image while the immense mountain landscape dominates the frame.
MOVEMENT:
Start already moving. She glides at a brisk, confident downhill speed, making broad, fluid S-shaped turns like the rider in the original video. Adapt the carving motion naturally to roller skates: softly bent knees, subtle weight shifts, alternating foot positions and relaxed arms balancing her body. Both skates roll convincingly on the asphalt. Her ponytail and jacket flutter backward.
The road and nearby grass sweep past quickly, while distant mountains move slowly with realistic parallax. Maintain steady forward speed, with no slow motion or abrupt acceleration.
SCENERY AND SURREAL STYLE:
Reproduce the original's extraordinary alpine valley: impossibly tall, jagged mountains covered in vivid green vegetation, towering rocky spires, conifer trees, colorful roadside flowers, a small village ahead and a rich cobalt-blue sky.
The upper landscape appears to stretch and fold overhead into an immense dreamlike canopy. Sculptural white clouds wrap around peaks at multiple heights, interwoven with long, vertically stretched wisps. Preserve this surreal spatial arrangement consistently as the camera advances.
Clouds drift and billow gently around the mountains. The changing view comes primarily from traveling through the world, with strong depth and continuous perspective. Preserve recognizable peaks, road geometry and village buildings throughout.
TIMING:
0–2 seconds: Immediate forward movement from the reference composition. The woman glides downhill near the right side of the road, then begins a gentle carve toward the middle of her lane. The camera follows smoothly.
2–4 seconds: A white car approaches in the opposite lane and passes on the left side of the frame, leaving a clear distance from the skater. She continues gliding naturally without reacting dramatically.
4–7 seconds: She makes a smooth S-shaped carve, shifting her weight and extending her arms slightly. The road's yellow center lines lead toward the village. Nearby trees and flowers move past, revealing more layers of the immense mountain valley.
7–10 seconds: She leans into the next gentle bend and continues downhill. The camera subtly follows the curve while preserving the rear view and full-body framing. Towering mountains and intricate clouds remain prominent overhead. End during continuous motion.
LOOK:
Match the original's vivid, richly textured surreal travel-video aesthetic: saturated alpine greens, deep blue sky, brilliant white clouds, strong natural sunlight and clear shadows. Photographic clothing, hair, asphalt and vegetation. Subtle peripheral motion blur, with the skater remaining sharp.
AUDIO:
Soft rushing wind and roller-skate wheels on asphalt. No dialogue.
AVOID:
Cuts, camera orbit, frontal views, dramatic zooms, cropping the feet, random scene changes, landscape melting, flickering clouds, floating feet, warped skates, extra wheels or limbs. No skateboard, bicycle, whales, portals, new fantasy creatures, text or logos.
Alex Neyef retweeted
⚠️If GTA 6 looks like this, I’ll buy it for the whole family.
Seedance 2.5 on @makeugc
Free prompt:
Create a 30-second, 16:9, ultra-photorealistic AAA third-person stealth video game gameplay sequence, inspired by GTA-style character movement, stealth mechanics, and interactive mission gameplay.
**CRITICAL: THIS MUST LOOK AND MOVE LIKE REAL VIDEO GAMEPLAY, NOT A MOVIE.**
### REFERENCE IMAGES
- **@Image1 — Female protagonist:** Preserve her exact facial appearance, short turquoise-blue bob haircut, silver hair clips, pink-and-white sporty outfit, pink backpack, body proportions, white socks, and sneakers.
- **@Image2 — Male boss:** Preserve his face, dark wavy hair, light stubble, muscular build, white open-collar shirt, dark trousers, and black shoes.
- ** — Storyboard:** Follow the scene composition, office environment, gameplay interface, and story progression. Use the updated coffee-machine distraction sequence and 30-second timing below.
### CAMERA AND GAMEPLAY — HIGHEST PRIORITY
The camera must behave like a third-person video game camera controlled by a player.
- Camera positioned approximately 2–3 meters behind the protagonist, slightly above shoulder height.
- Follow the character continuously during movement, maintaining consistent distance and framing.
- Show the protagonist's full body during walking, crouching, sneaking, and running.
- The camera smoothly rotates with her direction changes, just like a third-person action game.
- Movement must have realistic video-game animations: walk cycles, crouch-walking, stealth movement, running, turning, stopping, and interacting with objects.
- Natural foot placement, weight transfer, momentum, joint movement, and collision with the environment.
- No cinematic camera switching during normal gameplay.
- Brief scripted cutscenes are allowed only to show the boss's reaction, burger pickup, and final burger cutting.
- After each brief cutscene, immediately return to the third-person gameplay camera.
**IMPORTANT:** The protagonist must feel like a controllable 3D game character, not an actress being followed by a film camera.
### GAME INTERFACE
Persistent GTA-inspired game HUD:
- Health hearts and green health bar in the upper-left corner.
- Circular mini-map in the lower-left corner.
- Green stealth detection indicator.
- Mission objective text.
- Interactive button prompts.
- Red enemy detection alerts.
- Item acquisition notifications.
- Mission completion reward.
All interface text must be in English, clearly readable, and consistently positioned.
### TIMELINE — 30 SECONDS
**00:00–00:03 — MISSION START**
Third-person gameplay camera directly behind the female protagonist.
She enters a modern corporate office corridor, walking with a natural video-game character animation.
Her pink backpack moves subtly with her footsteps.
She holds a realistic sharp-pointed kitchen knife in her RIGHT hand, blade pointing downward.
The game HUD appears immediately:
"MAIN QUEST: STEAL YOUR LUNCH 🍔"
"DON'T GET CAUGHT"
The camera follows her from behind as she approaches the office area.
**00:03–00:05 — SPOT THE BOSS**
She approaches a wall and stops.
The character performs a realistic peek-around-corner animation.
The male boss sits at his desk working on a laptop.
A large, untouched cheeseburger sits on a plate on the BOSS'S DESK, beside his laptop.
The boss does not notice her.
HUD: "BOSS — DISTRACTED"
**00:05–00:08 — STEALTH MOVEMENT**
The protagonist transitions smoothly from standing into a realistic crouch.
She bends both knees naturally, lowers her center of gravity, and begins crouch-walking toward the kitchenette.
Her feet remain grounded.
Her knees, hips, and spine move with realistic biomechanics.
The third-person camera lowers slightly to follow her crouched position.
She moves between office furniture and partitions.
HUD: "STEALTH" with a green indicator.
No crawling, unnatural squatting, floating, sliding, or exaggerated poses.
**00:08–00:10 — COFFEE MACHINE DISTRACTION**
She reaches the coffee machine near the kitchenette entrance, away from the boss's desk.
She rises slightly from her crouch and presses the coffee machine START button with her LEFT hand.
The knife remains in her RIGHT hand, safely pointed downward.
HUD: "[E] START COFFEE"
The machine begins making a loud grinding and brewing noise.
She immediately moves behind a nearby partition.
**00:10–00:13 — THE BOSS INVESTIGATES**
The boss hears the loud coffee machine.
He looks away from his laptop, stands up naturally, and walks toward the kitchenette to investigate.
His cheeseburger remains on his desk.
HUD: "INVESTIGATING!"
A red exclamation mark appears above the boss.
A brief game-engine cutscene shows him leaving his desk.
**00:13–00:15 — WAITING IN COVER**
Return to third-person gameplay.
The protagonist remains hidden behind the partition.
She watches the boss walk past toward the coffee machine.
She waits until his back is turned and the path to his desk is clear.
Her crouched pose and movements remain realistic and balanced.
**00:15–00:18 — STEAL THE BURGER**
She quietly approaches the boss's unattended desk in crouch mode.
The camera follows her from behind.
The large cheeseburger is clearly visible on its plate next to the laptop.
HUD: "[E] PICK UP"
She reaches out with her LEFT hand, grips the intact burger, and lifts it completely off the plate.
Her RIGHT hand continues holding the same knife.
HUD notification:
"+1 BURGER"
"ITEM ACQUIRED"
The plate is now empty.
**00:18–00:21 — ESCAPE**
She stands up from her crouch and immediately starts running toward the office lounge.
Use an authentic third-person video game sprint animation.
The camera tracks behind her, slightly widening the field of view as she accelerates.
The burger remains CLEARLY VISIBLE in her LEFT hand.
The knife remains in her RIGHT hand, pointed downward.
Her backpack, hair, skirt, and limbs react naturally to running.
The mini-map displays her movement toward the lounge area.
**00:21–00:23 — DETECTED!**
The boss returns to his desk.
He notices the empty plate where the cheeseburger was.
He looks surprised and confused.
A red exclamation mark appears.
HUD: "DETECTED!"
A short scripted gameplay reaction shot shows the boss looking toward the corridor.
He does not teleport or suddenly appear near the protagonist.
**00:23–00:26 — REACH THE TABLE**
Return to third-person gameplay camera.
The protagonist reaches a lounge table and slows down.
She turns naturally toward a chair and performs a realistic sitting animation.
Her feet remain in contact with the floor, and her body weight shifts naturally into the chair.
She places the intact burger from her LEFT hand onto a clean plate on the table.
She prepares to cut it.
**00:26–00:29 — CUT THE BURGER**
A short, clear gameplay interaction close-up.
The protagonist holds the burger steady with her LEFT hand.
With her RIGHT hand, she uses the same sharp kitchen knife to cut the burger completely through the center.
Show the entire physical cutting action.
The knife blade contacts the top bun, moves downward through the ingredients, and reaches the plate.
The burger separates into exactly TWO realistic halves.
Visible beef patties, melted cheese, lettuce, and sauce.
No instant cutting, duplicated burgers, floating ingredients, or missing knife.
**00:29–00:30 — MISSION COMPLETE**
The protagonist looks down at the two burger halves and smiles with satisfaction.
The camera remains in the game environment.
A green checkmark appears with the final HUD notification:
"MISSION COMPLETE"
"LUNCH SECURED"
"+100 XP"
End on a satisfying gameplay victory screen.
### STRICT CONTINUITY RULES
1. Maintain the exact appearances of @Image1 and @Image2 throughout the video.
2. Follow for the storyboard's visual style and overall office environment, but follow this updated timeline for the coffee-machine sequence.
3. The burger is ALWAYS on the boss's desk until stolen, never beside the coffee machine.
4. The coffee machine distracts the boss and causes him to leave his desk.
5. The protagonist waits until the boss has walked away before approaching his desk.
6. The protagonist carries the knife in her RIGHT hand and the stolen burger in her LEFT hand.
7. The burger remains visibly in her LEFT hand during the entire escape.
8. The protagonist places the burger on the plate before cutting it.
9. Only one burger exists until it is physically cut into two halves.
10. All character movement must resemble physically believable third-person game animations.
11. No teleportation, object disappearance, hand switching, outfit changes, identity changes, or unrealistic crouching.
12. No spoken dialogue.
### VISUAL AND AUDIO STYLE
Ultra-realistic 3D game graphics, physically based rendering, natural indoor lighting, detailed office textures, smooth character animation, subtle motion blur, responsive game camera, realistic collision physics.
Sound design includes footsteps, clothing movement, office ambience, coffee grinding, interactive pickup sound, alert notification, and mission-complete chime.
**FINAL PRIORITY: The result must feel like a continuous, playable third-person stealth mission recorded directly from a modern AAA video game, with believable character controls and interactions — NOT a cinematic film or music video.
🤝 Paid partnership · 🤖 Made with AI
⚠️This is what games will look like when AI takes over
Does anyone know any top Asian games? I really like the visual.
Seedance 2.5 on @makeugc
PROMPT:
Create a 30-second photorealistic stealth-game video in 16:9.
REFERENCE PRIORITY
is the authoritative reference for the adult female protagonist: preserve her face, hairstyle, physique, outfit, footwear and backpack throughout.
-Image2 is the storyboard for the first half.
-Image3 is the storyboard for the second half.
Read each storyboard from left to right, top to bottom. Follow all six panels of Image2, then all six panels of Image3.
The storyboard images define actions, locations, camera compositions and game interface. If the protagonist’s appearance differs between references, Image1 takes priority.
IMPORTANT
Produce full-screen moving scenes. The storyboard grids are planning references only: never display panels, borders, collages or split screens. Animate the actions between the illustrated moments naturally rather than presenting a slideshow.
VISUAL STYLE
Realistic third-person stealth gameplay with cinematic camera work, believable human motion and natural daylight. Modern university athletics administration wing: sage-green walls, pale terrazzo floors, oak doors with glass panels, black-framed windows and indoor plants.
Maintain consistent room geography: the office entrance connects to the corridor, the sage-green drawer cabinet stands against the side wall, and the solid-front oak desk stands near the windows. Preserve these positions across camera changes.
The protagonist wears her backpack on BOTH shoulders throughout. Both straps remain attached and the backpack moves naturally with her body.
SHOT SEQUENCE
0–3 seconds:
Start immediately with the protagonist concealed beside a corridor corner. The camera looks past her shoulder toward a male staff member walking away. She cautiously peeks around the corner and waits until he disappears from view.
HUD status: “STAY HIDDEN”, then “WAIT FOR PATROL”.
3–6 seconds:
She approaches the office door and looks through its glass panel. Show her checking the empty office before touching the handle. She listens briefly, then reaches for the handle.
HUD status: “CHECK THE OFFICE”.
6–9 seconds:
Close view of her hand pressing the handle, followed by a medium shot as she opens the door and slips inside. The camera follows smoothly through the doorway. She leaves the door slightly ajar.
HUD status: “OPEN DOOR”, then “ENTER QUIETLY”.
9–12 seconds:
She moves to the sage-green cabinet, crouches beside it and pulls out a drawer. Show the drawer sliding along its rails and her hands searching inside.
HUD status: “SEARCH DRAWER”.
12–15 seconds:
She spots her pink-cased smartphone, grasps it and lifts it from the drawer. The phone is clearly visible. Her expression briefly shows relief. She quietly pushes the drawer closed.
HUD status: “PHONE RECOVERED”.
Change the main objective to “ESCAPE UNSEEN”.
15–18 seconds:
Footsteps approach. She freezes, turns toward the doorway and lowers herself behind the cabinet before a female lecturer enters. The lecturer has a short dark bob, a teal blazer and cream trousers, matching Image3. She carries folders and walks toward the bookcase without noticing the protagonist.
HUD warning: “TEACHER RETURNING”.
The ALERT meter rises to amber because of the nearby threat, without reaching full detection.
18–21 seconds:
While the lecturer faces away, the protagonist moves into concealment behind the solid-front oak desk. Use a clear wide shot to establish that the desk blocks the lecturer’s line of sight. She holds the recovered phone close and watches the lecturer carefully.
HUD status: “STAY OUT OF SIGHT”.
21–25 seconds:
The lecturer remains turned toward the bookcase. The protagonist quietly moves from behind the desk toward the office exit, staying outside the lecturer’s field of view. The camera tracks her movement smoothly. She crosses the doorway into the corridor.
HUD status: “MOVE TO THE EXIT”.
25–28 seconds:
From the corridor, she gently pulls the office door closed. The latch clicks softly. She glances back, then relaxes as the ALERT meter falls to zero.
HUD status: “CLOSE THE DOOR”.
28–30 seconds:
Medium frontal tracking shot as she walks down the corridor. She raises the recovered phone at chest height and gives a small relieved smile. Keep both backpack straps visible.
Main objective changes to “MISSION COMPLETE”.
Final status: “PHONE RECOVERED”.
GAME INTERFACE
Match the HUD design shown in Image2 and Image3: small translucent charcoal panels, thin pale outlines, cyan accents and pink or amber alert indicators.
Keep the interface anchored to the screen, with identical placement and typography across every shot:
- Top left: “HEALTH”, “STAMINA”, “ALERT”.
- Top center: the current mission objective.
- Bottom left: a circular mini-map showing the protagonist’s position and orientation.
- Bottom right: “C CROUCH”, “E INTERACT”, “SHIFT SPRINT”.
- Lower center: the current action or warning.
Initial objective: “RECOVER YOUR PHONE”.
After collecting the phone: “ESCAPE UNSEEN”.
Final objective: “MISSION COMPLETE”.
HEALTH stays constant. STAMINA changes subtly with movement. ALERT responds to the lecturer’s proximity and returns to zero after escape.
Keep all English text sharp, correctly spelled and stable. Never cover the protagonist’s face or hands with interface elements.
AUDIO
Quiet corridor ambience, distant footsteps, soft sneaker steps, backpack movement, door-handle click, drawer sliding and a restrained suspense soundtrack.
A subtle notification sound when the phone is recovered, a brief warning tone when the lecturer approaches, and a short success chime at the end.
No dialogue or voice-over.
CONTINUITY
Natural movement, realistic weight and hand contact with objects. No teleportation, duplicate characters, extra limbs or changing outfits. The phone remains in her hand after collection. Doors and drawers move physically and consistently.
The lecturer never directly sees the protagonist.
Maintain a playful, tense stealth-game atmosphere and a complete beginning, middle and ending.
Alex Neyef retweeted
This is just insane. I asked AI to bring my 6-year-old daughter’s drawings to life.🥰
Now I was able to see this world through my daughter’s eyes.
Seedance 2.5 on @makeugc
Children see a much broader picture than we can even imagine. Think back to your own childhood - all the colors were so much more vivid.
I had seen something similar online, but I never imagined it could turn out this cool.
🤝 Paid partnership · 🤖 Made with AI
⚠️If GTA 6 looks like this, I’ll buy it for the whole family.
Seedance 2.5 on @makeugc
Free prompt:
Create a 30-second, 16:9, ultra-photorealistic AAA third-person stealth video game gameplay sequence, inspired by GTA-style character movement, stealth mechanics, and interactive mission gameplay.
**CRITICAL: THIS MUST LOOK AND MOVE LIKE REAL VIDEO GAMEPLAY, NOT A MOVIE.**
### REFERENCE IMAGES
- **@Image1 — Female protagonist:** Preserve her exact facial appearance, short turquoise-blue bob haircut, silver hair clips, pink-and-white sporty outfit, pink backpack, body proportions, white socks, and sneakers.
- **@Image2 — Male boss:** Preserve his face, dark wavy hair, light stubble, muscular build, white open-collar shirt, dark trousers, and black shoes.
- ** — Storyboard:** Follow the scene composition, office environment, gameplay interface, and story progression. Use the updated coffee-machine distraction sequence and 30-second timing below.
### CAMERA AND GAMEPLAY — HIGHEST PRIORITY
The camera must behave like a third-person video game camera controlled by a player.
- Camera positioned approximately 2–3 meters behind the protagonist, slightly above shoulder height.
- Follow the character continuously during movement, maintaining consistent distance and framing.
- Show the protagonist's full body during walking, crouching, sneaking, and running.
- The camera smoothly rotates with her direction changes, just like a third-person action game.
- Movement must have realistic video-game animations: walk cycles, crouch-walking, stealth movement, running, turning, stopping, and interacting with objects.
- Natural foot placement, weight transfer, momentum, joint movement, and collision with the environment.
- No cinematic camera switching during normal gameplay.
- Brief scripted cutscenes are allowed only to show the boss's reaction, burger pickup, and final burger cutting.
- After each brief cutscene, immediately return to the third-person gameplay camera.
**IMPORTANT:** The protagonist must feel like a controllable 3D game character, not an actress being followed by a film camera.
### GAME INTERFACE
Persistent GTA-inspired game HUD:
- Health hearts and green health bar in the upper-left corner.
- Circular mini-map in the lower-left corner.
- Green stealth detection indicator.
- Mission objective text.
- Interactive button prompts.
- Red enemy detection alerts.
- Item acquisition notifications.
- Mission completion reward.
All interface text must be in English, clearly readable, and consistently positioned.
### TIMELINE — 30 SECONDS
**00:00–00:03 — MISSION START**
Third-person gameplay camera directly behind the female protagonist.
She enters a modern corporate office corridor, walking with a natural video-game character animation.
Her pink backpack moves subtly with her footsteps.
She holds a realistic sharp-pointed kitchen knife in her RIGHT hand, blade pointing downward.
The game HUD appears immediately:
"MAIN QUEST: STEAL YOUR LUNCH 🍔"
"DON'T GET CAUGHT"
The camera follows her from behind as she approaches the office area.
**00:03–00:05 — SPOT THE BOSS**
She approaches a wall and stops.
The character performs a realistic peek-around-corner animation.
The male boss sits at his desk working on a laptop.
A large, untouched cheeseburger sits on a plate on the BOSS'S DESK, beside his laptop.
The boss does not notice her.
HUD: "BOSS — DISTRACTED"
**00:05–00:08 — STEALTH MOVEMENT**
The protagonist transitions smoothly from standing into a realistic crouch.
She bends both knees naturally, lowers her center of gravity, and begins crouch-walking toward the kitchenette.
Her feet remain grounded.
Her knees, hips, and spine move with realistic biomechanics.
The third-person camera lowers slightly to follow her crouched position.
She moves between office furniture and partitions.
HUD: "STEALTH" with a green indicator.
No crawling, unnatural squatting, floating, sliding, or exaggerated poses.
**00:08–00:10 — COFFEE MACHINE DISTRACTION**
She reaches the coffee machine near the kitchenette entrance, away from the boss's desk.
She rises slightly from her crouch and presses the coffee machine START button with her LEFT hand.
The knife remains in her RIGHT hand, safely pointed downward.
HUD: "[E] START COFFEE"
The machine begins making a loud grinding and brewing noise.
She immediately moves behind a nearby partition.
**00:10–00:13 — THE BOSS INVESTIGATES**
The boss hears the loud coffee machine.
He looks away from his laptop, stands up naturally, and walks toward the kitchenette to investigate.
His cheeseburger remains on his desk.
HUD: "INVESTIGATING!"
A red exclamation mark appears above the boss.
A brief game-engine cutscene shows him leaving his desk.
**00:13–00:15 — WAITING IN COVER**
Return to third-person gameplay.
The protagonist remains hidden behind the partition.
She watches the boss walk past toward the coffee machine.
She waits until his back is turned and the path to his desk is clear.
Her crouched pose and movements remain realistic and balanced.
**00:15–00:18 — STEAL THE BURGER**
She quietly approaches the boss's unattended desk in crouch mode.
The camera follows her from behind.
The large cheeseburger is clearly visible on its plate next to the laptop.
HUD: "[E] PICK UP"
She reaches out with her LEFT hand, grips the intact burger, and lifts it completely off the plate.
Her RIGHT hand continues holding the same knife.
HUD notification:
"+1 BURGER"
"ITEM ACQUIRED"
The plate is now empty.
**00:18–00:21 — ESCAPE**
She stands up from her crouch and immediately starts running toward the office lounge.
Use an authentic third-person video game sprint animation.
The camera tracks behind her, slightly widening the field of view as she accelerates.
The burger remains CLEARLY VISIBLE in her LEFT hand.
The knife remains in her RIGHT hand, pointed downward.
Her backpack, hair, skirt, and limbs react naturally to running.
The mini-map displays her movement toward the lounge area.
**00:21–00:23 — DETECTED!**
The boss returns to his desk.
He notices the empty plate where the cheeseburger was.
He looks surprised and confused.
A red exclamation mark appears.
HUD: "DETECTED!"
A short scripted gameplay reaction shot shows the boss looking toward the corridor.
He does not teleport or suddenly appear near the protagonist.
**00:23–00:26 — REACH THE TABLE**
Return to third-person gameplay camera.
The protagonist reaches a lounge table and slows down.
She turns naturally toward a chair and performs a realistic sitting animation.
Her feet remain in contact with the floor, and her body weight shifts naturally into the chair.
She places the intact burger from her LEFT hand onto a clean plate on the table.
She prepares to cut it.
**00:26–00:29 — CUT THE BURGER**
A short, clear gameplay interaction close-up.
The protagonist holds the burger steady with her LEFT hand.
With her RIGHT hand, she uses the same sharp kitchen knife to cut the burger completely through the center.
Show the entire physical cutting action.
The knife blade contacts the top bun, moves downward through the ingredients, and reaches the plate.
The burger separates into exactly TWO realistic halves.
Visible beef patties, melted cheese, lettuce, and sauce.
No instant cutting, duplicated burgers, floating ingredients, or missing knife.
**00:29–00:30 — MISSION COMPLETE**
The protagonist looks down at the two burger halves and smiles with satisfaction.
The camera remains in the game environment.
A green checkmark appears with the final HUD notification:
"MISSION COMPLETE"
"LUNCH SECURED"
"+100 XP"
End on a satisfying gameplay victory screen.
### STRICT CONTINUITY RULES
1. Maintain the exact appearances of @Image1 and @Image2 throughout the video.
2. Follow for the storyboard's visual style and overall office environment, but follow this updated timeline for the coffee-machine sequence.
3. The burger is ALWAYS on the boss's desk until stolen, never beside the coffee machine.
4. The coffee machine distracts the boss and causes him to leave his desk.
5. The protagonist waits until the boss has walked away before approaching his desk.
6. The protagonist carries the knife in her RIGHT hand and the stolen burger in her LEFT hand.
7. The burger remains visibly in her LEFT hand during the entire escape.
8. The protagonist places the burger on the plate before cutting it.
9. Only one burger exists until it is physically cut into two halves.
10. All character movement must resemble physically believable third-person game animations.
11. No teleportation, object disappearance, hand switching, outfit changes, identity changes, or unrealistic crouching.
12. No spoken dialogue.
### VISUAL AND AUDIO STYLE
Ultra-realistic 3D game graphics, physically based rendering, natural indoor lighting, detailed office textures, smooth character animation, subtle motion blur, responsive game camera, realistic collision physics.
Sound design includes footsteps, clothing movement, office ambience, coffee grinding, interactive pickup sound, alert notification, and mission-complete chime.
**FINAL PRIORITY: The result must feel like a continuous, playable third-person stealth mission recorded directly from a modern AAA video game, with believable character controls and interactions — NOT a cinematic film or music video.
Alex Neyef retweeted
Made with Seedance 2.5 on @makeugc using a video reference
PROMPT 👇
Edit Video 1. Replace ONLY the original person with MILA as described below, and expand the video to a horizontal 16:9 frame. Preserve the original action, timing, environment and camera movement.
CHARACTER
A beautiful American Instagram-style brunette model and performer, 20 years old. Long dark-brown hair in a high ponytail, glossy lips, expressive eyes, soft glam makeup and delicate sparkly stud earrings. Slim athletic figure with a very large, full bust, very full rounded buttocks, a narrow waist, flat stomach and lean arms and legs.
Replace the original person’s entire appearance, including head, face, hair, skin and body, consistently throughout the video. Use described proportions while preserving the original person’s poses and movement.
OUTFIT
MILA wears a stylish cherry-red satin bodycon mini dress with thin straps, a structured supportive bodice, a deep sweetheart neckline showing cleavage, a fitted waist and a mid-thigh hem. Bare shoulders, arms and legs. Realistic satin texture, natural folds and fabric movement. The dress remains opaque and securely fitted throughout the action.
CAST AND CRUTCHES
Keep the same white cast around the same foot and ankle as in the source video. Match its coverage, shape and position; do not extend it up the leg or switch it to the opposite foot. Keep the other foot bare.
Preserve both original crutches, including their design, size, hand grips, ground contact and movement. Match the original hand placement and the exact moments when each crutch is lifted, released and thrown aside. Once discarded, each crutch stays where the source action places it.
ACTION AND CONTINUITY
Use Video 1 as the exact motion reference. Preserve the full sequence: approaching the water on crutches, discarding the crutches one after the other, balancing and hopping at the edge, losing balance and falling into the water.
Match the original timing, body orientation, gestures, foot placements, trajectory of the fall and final splash. Preserve the original back and three-quarter views. Do not introduce new turns toward the camera, poses or close-ups to show MILA’s face.
Keep the muddy water, sloping bank, dry bamboo leaves, bamboo grove, visible thin line across the scene and all other environmental details consistent with the source.
16:9 COMPOSITION
Create a true horizontal 16:9 image by extending the environment naturally to the left and right of the original frame. Preserve the original vertical field of view, camera perspective and subject size relative to frame height. Continue the bank, water and bamboo grove coherently across the added areas, with stable geometry throughout the camera movement.
Do not crop away the head, feet, crutches or fall. Do not stretch the original image, add black bars or use blurred side panels.
REALISM AND OUTPUT
Match the source’s natural daylight, shadows, handheld movement, focus and image texture. She must look like a real person filmed in this exact location, with convincing contact shadows and natural interaction between her hands, crutches, feet, clothing and surroundings.
Preserve the original 19.8-second duration, playback speed, frame rate and audio. No added dialogue, music, cuts, captions or extra people.
🤖 Made with AI
Somehow ended up with budget Dua Lipa, but the video turned out pretty good lol 😂
Made with Seedance 2.5 on @makeugc using a long text prompt. The ONLY reference was the fictional candy packaging. No character references, the model came up with her itself.
The first 2–3 attempts got blocked by moderation, so I had to retry. No idea why. Maybe it was trying to make Dua Lipa even though I never asked for her. The version that finally went through gave me this half-Dua situation lol.
Then I added the music and effects with AI.
PROMPT in the comments 👇 It’s a long one.
#AIVideo














