← Seedance 2.5 prompt guide

Performance and emotion

Quiet scenes carried by gaze, hands, props and distance rather than by adjectives.

The templateAll 36 clips

Toddler mischief in three rooms

scene list · Seedance 2.5 per the author · Oct 28, 2026

Seven short scenes escalate one joke (water in the bathroom, a spilled glass in the kitchen, a puddle outside) and end on a freeze-frame payoff. Each character's look is written out once in a consistency block, down to shirt pattern and hair, and the mother's reaction gets its own close-up scene. There are no timecodes or stated length, so pacing is left to the model.

Performance and emotion

Read @Zarnab_with_Ai's prompt
Create a high-quality 3D animated family comedy video featuring an adorable mischievous toddler boy and his young mother. Use a polished Pixar-inspired animation style, expressive facial features, soft cinematic lighting, realistic water physics, detailed environments, and smooth character animation. Vertical 9:16 aspect ratio, 4K UHD quality, 24fps.

Scene 1 – Bathroom Mischief: A cute toddler boy with messy brown hair, huge expressive brown eyes, a yellow T-shirt with small white cloud patterns, and navy-blue shorts is playing with water in a modern bathroom. The floor is wet and covered with splashes. His young mother, wearing a dusty rose T-shirt and light gray shorts, enters carrying a laundry basket and discovers the mess. The toddler looks innocent and mischievous.

Scene 2 – Mother's Reaction: Show a close-up of the mother's expressive face as she raises her eyebrows in surprise and looks at her naughty little son. Use cinematic close-up framing and exaggerated but natural animated expressions.

Scene 3 – Cleaning Up: The mother gently picks up the toddler and tries to clean him up near the toilet. The little boy squirms playfully, making it difficult for her to manage him. Show realistic body movement, affectionate parenting, and humorous timing.

Scene 4 – The Glass of Water: Cut to a warm, sunlit kitchen with wooden countertops, large windows, and soft curtains. The mother places a transparent glass filled with water on the kitchen table and steps away briefly. The toddler notices the glass and curiously approaches it.

Scene 5 – Unexpected Splash: The toddler climbs toward the table, accidentally spills the water, and slides onto the wet tabletop. Water spreads across the surface as he crawls and giggles. Use detailed water splashes, realistic reflections, and playful cartoon physics.

Scene 6 – Mom Rushes In: The mother hurries into the kitchen, lifts her mischievous toddler into her arms, and carries him toward the door. Her expression shows exhaustion, surprise, and disbelief at his endless mischief.

Scene 7 – Outdoor Chaos: The mother brings the toddler outside. He suddenly runs toward a shallow puddle on the wet ground. She reaches out to catch him, but he jumps into the water with a huge joyful splash. Water droplets fly dramatically in every direction.

Final Shot: Capture the toddler laughing happily in the puddle while his mother stands behind him, completely shocked, with her hair standing wildly on end from the enormous splash. The toddler remains delighted and innocent. End with a funny cinematic freeze-frame.

Visual Style: Premium 3D animated family film, Pixar-inspired character design, oversized expressive eyes, soft skin shading, detailed hair, warm natural lighting, cinematic depth of field, realistic water simulation, smooth camera transitions, wide shots mixed with expressive close-ups, comedic timing, and consistent character design.

Character Consistency: Maintain the exact same toddler face, hairstyle, yellow cloud-pattern shirt, navy shorts, and body proportions in every scene. Keep the mother’s face, brown hair in a messy bun, dusty rose T-shirt, and gray shorts consistent throughout the video.

Audio: Playful instrumental background music, water splashing sounds, toddler giggles, and subtle comedic sound effects. No dialogue, no subtitles, and no text overlays.

Original post ↗

What holds it together

Seedance 2.5 handles intimate and comic scenes when the prompt gives it behaviour to film rather than feelings to display. The best examples stage emotion through what people do: a drink offered and taken back, a hand reaching into frame, a pause before standing, a deadpan paw lick after a knockout. A breakup scene in the library defines one eyeline axis, keeps the camera on one side of it, and uses an arriving train's headlight as a light source that grows with the tension. Amateur formats such as vlogs, phone clips and VHS work when their flaws are written as camera behaviour (focus hunting, late framing) instead of as quality words. What goes wrong: action-style constant motion imported into a quiet scene, emotion adjectives with no behaviour behind them, crossing the axis, and props that vanish between beats.

Write these in

  1. The emotional arc as a sequence of small actions, one per beat
  2. Props that carry the relationship and stay continuous
  3. One eyeline axis and the side of it the camera stays on
  4. Short dialogue lines placed inside specific beats
  5. Camera behaviour that fits the format: partner's handheld, phone, documentary
  6. What the scene must not become: no melodrama, no posing, no forced kiss
  7. For physical comedy, the source of every movement and what must not happen before each step

What breaks it

  • Writing 'she feels heartbroken'. Write what her hands and eyes do instead.
  • Using fight-style density. One look, one hesitation and one reach can fill five seconds.
  • Asking for amateur footage as a quality level. Describe the operator's mistakes.
  • Losing props. A cup set down stays on the bench.

The template

[DURATION]-second, 16:9, [live-action / animated] [scene type]. [CHARACTER A: look, outfit]. [CHARACTER B: look, outfit]. Faces and outfits stay identical throughout.
Setting: [location, time, light source that changes during the scene: e.g. approaching headlight, setting sun].
Axis: A stays screen-left looking right, B screen-right looking left; the camera stays on [side] of that line.
Props: [prop 1] with A, [prop 2] between them. They stay where they are put.
0-[X]s: [A's small action], [B's response]. Close on [hands / eyes]. Line: '[short line]'.
[X]-[Y]s: [turn in the relationship shown through an action: standing, handing over, stepping closer].
[Y]-[Z]s: [the key gesture]. The [light source] reaches [state].
[Z]-end: [resolution action]. Hold on [final image].
Not: [melodrama, slow-motion running, exaggerated tears, posing to camera].

Ours to use freely. Replace everything in brackets.

36 clips that use it

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

Penguin, giant apple and a bark sled

Shot-by-shot timeline · Seedance 2.5 per the author · Oct 25, 2026

Most of the prompt is a physics ledger: one apple, one bark slab, which side is uphill, and when the penguin may touch, push or ride the bark (never before 0:15.5). Every movement is given a source (his feet for pulling and pushing, gravity only once the slope is visible, leaf friction to stop), and the punchline depends on the apple being wider than the doorway from the first frame. Framing rules ask for wide side views wherever cause and effect matters.

Performance and emotionChases, racing and fast movement

Sleeping guard beats five thieves

Shot-by-shot timeline · Seedance 2.5 per the author · Oct 21, 2026

The joke is a constraint held for 30 seconds: she stays asleep, and every defensive move is written as an unconscious movement such as a stray arm, an outstretched leg or an elbow. Thieves drop out one at a time and stay down, so the count runs from five to zero by the 18-second mark and leaves the rest for the police reaction. Short mumbled lines are placed inside specific shots, and the style block asks for harmless physical gags.

Performance and emotionClose combat and impact

Read @doctorwasif's prompt
Create a cinematic, ultra-realistic slapstick comedy scene. The watchwoman’s face and outfit must match the reference image throughout.

Shot 1 (0–3s): At a warehouse entrance, a night watchwoman sleeps deeply in a chair, feet on a crate, snoring with a half-eaten bag of chips on her lap. Five thieves sneak in, smirking at her.

Shot 2 (3–8s): The thieves try passing her, but her unconscious movements effortlessly clothesline one, trip another into crates, and send a third crashing into a shelf. She keeps eating chips and sleepily mutters, “Too slow…”

Shot 3 (8–13s): The remaining thieves nervously huddle before the leader charges her. Without opening her eyes, she elbows him in the stomach, mumbling, “Five more minutes.” He collapses beside the others.

Shot 4 (13–18s): The last thief attempts to escape, but her outstretched leg trips him into boxes. All five thieves lie scattered across the warehouse while she finishes her chips, tosses the empty bag perfectly into a trash can, and uses one thief as a footrest.

Shot 5 (18–23s): Police sirens approach. She briefly opens one eye, surveys the defeated thieves with a deadpan expression, mutters, “Called it,” and immediately falls back asleep. Her drool resumes as she snores.

Shot 6 (23–30s): Police rush in, find the thieves defeated, and stare at the sleeping watchwoman in disbelief. An officer tries waking her, but she lazily swats his hand away. The final close-up reveals a fresh bag of chips mysteriously resting in her lap as she smiles peacefully in her sleep.

Style: Fast-paced slapstick comedy, exaggerated but harmless physical gags, expressive reactions, consistent character appearance, smooth shot transitions, cinematic warehouse lighting, impeccable comedic timing, and a satisfying punchline.

Original post ↗

A zombie and a girl fight over a snack bag

shot list · Seedance 2.5 per the author · Oct 20, 2026

Four reference images are named inline at the beat each one governs: the opening frame, the snack, the zombie reveal and the final two-shot. The scare is built in horror order (a hand on the shoulder, a freeze, a slow head turn) before the scene flips to comedy and ends on a product close-up. The tug-of-war is only asked to be long and funny, so that stretch is the least specified part of the prompt.

Performance and emotionHorror, suspense and POVProduct, cars and trailers

Read @harboriis's prompt
The opening frame is Image 1
and the snack the girl is eating is Image 2

From 0–0.5 seconds, use a medium shot to establish the setting: the girl is sitting alone on the sofa in a dimly lit living room. The only light in the room comes from the television, which is playing a Resident Evil zombie movie. The camera then moves in closer, showing her clutching a large bag of snacks. She watches the movie nervously while mechanically stuffing snacks into her mouth.
Image 3
Cut to a close-up of the girl. A pale, blood-stained hand quietly rests on her shoulder, slowly reaching out from behind her and directly into the snack bag in her arms. Her entire body freezes. She slowly turns her head.
Reverse angle: a zombie is standing behind her. Once the girl comes to her senses, she immediately hugs the snack bag tightly. The monster refuses to give in and reaches over to snatch it away. This tug-of-war over the snacks should last for a long time and be very funny.
Suddenly cut to a medium shot: Image 4
 the two of them are now sitting side by side on the sofa, watching television and eating snacks together. The camera zooms in on a close-up of the product.

Original post ↗

Giant cat wins a slap contest

Shot-by-shot timeline · Seedance 2.5 per the author · Sep 3, 2026

Phone-footage flaws are written as camera behaviour (digital zoom, focus hunting), and the beats are timed for comedy: hook, tension, one slap, a deadpan paw lick. Filmer lines sit inside specific beats.

Performance and emotionClose combat and impact

Family animation trailer: the girl who bakes the sun

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 31, 2026

Named characters each get a one-line design, and a voice-over line per beat carries a full story arc in 30 seconds. Trailer structure (aerial reveal, comic beat, threat, hero moment) gives the model clear tonal shifts.

Performance and emotionAtmosphere and realismGiants, armies and epic scaleEnergy, spells and blasts

Read @CharaspowerAI's prompt
Premium stylized family-animation trailer with high-end feature-film quality, luminous color design, expressive faces, soft global illumination, rich volumetric clouds, cinematic depth and emotional orchestral trailer energy. Dynamic camera with sweeping aerials, playful close-ups, fast push-ins and epic hero reveals. Original premium animated-feature look, warm, magical and heartfelt.

Main hero: Nova, a brave 11-year-old girl with huge expressive eyes, curly dark hair, soot-smudged cheeks, oversized baker-apprentice coat and a glowing fire-key around her neck. Sidekick: Crumb, a tiny grumpy puff-dragon made of ash and embers, comic and lovable. Antagonistic force: a vast creeping living darkness swallowing the sky.

A fantastical sky bakery above the clouds, giant ovens forging each sunrise, floating flour mills, rope bridges, glowing dawn dough, stormy sky canyons and a final radiant sunrise tower.

0-5s: Epic aerial reveal - Vast cloud kingdom at dawn. Giant ovens glow above the sky. Voice-over in English: “Every morning… someone had to bake the sun.”

5-10s: Comedy + character hook - Nova rushes through the bakery carrying a glowing sunrise loaf. Crumb sneezes sparks. One tiny mistake. The glowing loaf slips. The entire sky goes dark. Record-scratch trailer beat.

10-16s: Adventure montage - Fast cuts: Nova flying on a flour sack through cloud canyons, Crumb screaming, giant gears turning, floating mills collapsing, Nova grabbing the fire-key. Voice-over: “And this year… it was her turn to fix it.”

16-22s: Threat escalation - The living darkness spreads across the horizon, swallowing floating islands. Nova stands against the storm, fire-key glowing in her hand. Crumb gulps.

22-27s: Hero climax - Slow-motion close-up on Nova’s determined face. She leaps toward the sunrise tower and thrusts the fire-key into its heart. Golden light erupts through the clouds.

27-30s: Trailer ending - Massive dawn floods the sky-world. Hard cut to Nova and Crumb covered in soot, laughing. Voice-over: “This fall… rise with the light.” Final title card: THE GIRL WHO BAKED THE DAWN.

Original premium animated-feature quality, strong facial acting, beautiful cloud simulation, trailer pacing, emotional color scripting, stylized but believable motion, no text except final title card.

Original post ↗

Last-train breakup on a subway platform

Shot-by-shot timeline · Seedance 2.5, inferred · Aug 18, 2026

A seven-step emotional arc, a single eyeline axis and props left as traces of the night carry the drama. The approaching train's headlight grows from a dot to a backlight as the tension rises.

Performance and emotionAtmosphere and realismProduct, cars and trailers

POV street invaded by 2D cartoons

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 17, 2026

Two rule sets coexist: real environments with natural light, and animated characters with bold outlines and elastic timing. A route from street to hall to backstage gives the POV somewhere to go.

Performance and emotion

Read @CharaspowerAI's prompt
POV sequence in an original live-action and hand-drawn 2D animation hybrid world. The entire scene is filmed from first-person perspective with smooth handheld movement, quick pans and playful reframing. Real environments use natural lighting and realistic shadows; animated characters use bold ink outlines, flat colors, elastic motion and expressive cartoon timing. Retro urban comedy, fully original design, no recognizable characters, no studio references.

A fictional old-fashioned city street leading into a colorful entertainment hall, then a narrow backstage corridor and exterior alley.

0-5s: POV moves through a lively street. Real pedestrians walk normally while small abstract cartoon creatures bounce between signs, stretch across doorways and leave hand-drawn motion trails.

5-10s: POV enters the hall. A tiny animated server slides across a real counter carrying wobbling drinks, spins once, waves at camera and disappears behind a curtain.

10-16s: POV turns toward a live performer on stage while animated musicians play around them. One cartoon instrument stretches toward the lens, curls into a spiral, then snaps gently back into place.

16-22s: POV backs away as playful animated chaos spreads through the room: drawn paper shapes float, a cartoon character flattens under a door, another leaves a cloud of sketched stars. Real people remain completely serious.

22-27s: POV moves quickly down the corridor while a small animated guide runs beside camera and says in English: “Keep moving. The drawings are improvising again.”

27-30s: POV exits into the alley. A huge inflatable cartoon prop swings past the lens, bounces harmlessly off a trash bin and launches colorful drawn stars into the air. End on a playful freeze-frame.

Original post ↗

Lover's-eye travel vlog

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 11, 2026

Four images define one identity (face, hair, build only), and the camera is the unseen partner's handheld. Intimacy comes from offering a drink, holding hands and sitting close rather than from posed shots.

Performance and emotionAtmosphere and realism

Staged tabloid leak at a waterfront villa

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 6, 2026

The camera hides behind leaves and a balcony rail and struggles with zoom and focus. The prompt specifies fictional consenting adults and no resemblance to real people.

Performance and emotion

Read @CharaspowerAI's prompt
Ultra-realistic low-quality tabloid footage from a staged paparazzi-style scene. Old handheld consumer camera with cheap long zoom, heavy compression, unstable autofocus, rolling shutter, blown highlights, digital noise, accidental reframing and muffled wind audio. Bright humid Miami afternoon. Raw amateur look, no cinematic polish.

Authentic low-grade tabloid texture, fictional adults, staged consent, imperfect focus, realistic sunlight and water reflections, natural body proportions, no nudity, no logos, no readable text.

One fictional adult female celebrity, glamorous, wearing an elegant swim outfit and sunglasses. One fictional adult man in an open linen shirt and swim shorts. Both are consenting performers in a staged scene, with no resemblance to real public figures.

Luxury Miami villa terrace with infinity pool, palm trees, white stone, glass walls and ocean skyline. Camera hidden behind leaves and a distant balcony railing.

0-5s: Shaky extreme long shot - The camera searches through palm leaves, loses focus, then catches the woman walking barefoot beside the pool. The operator whispers: “Wait… is that her?”
5-10s: Unstable digital zoom -  The man approaches. They laugh, stand close and exchange a brief romantic kiss. Autofocus pulses between leaves and their faces.
10-15s: Crash zoom + bad reframing - She suddenly looks directly toward the lens. The operator lowers the camera in panic, showing only the railing, pool reflections and part of the terrace.
15-20s: Messy handheld recovery -  The camera rises again. Both performers are now staring straight at the lens. The woman smiles, waves and says: “You really thought this was leaked?” Abrupt cut to black.

Original post ↗

Medieval market frozen by a gesture

Continuous action chain · Seedance 2.5 per the author · Aug 5, 2026

One paragraph describes a continuous walk through a paused market with a list of small interactions: an apple taken, bread straightened, a coin returned. The camera moves backward, circles and ends on a close-up.

Performance and emotion

Read @CharaspowerAI's prompt
ultra-realistic medieval market scene in daylight, filmed as one smooth continuous Steadicam shot.

A confident medieval traveler wearing a rough brown wool tunic, leather belt and worn boots walks slowly through a small village market. Around him are simple wooden stalls, baskets of apples, bread, cloth awnings, two chickens and one calm donkey beside a cart. A few villagers move naturally in the background.

After several seconds, the traveler makes a small hand gesture. The entire market suddenly becomes perfectly motionless, like a paused film. Villagers stop in place, the chickens become astill, the donkey stops moving, cloth awnings stop in the wind.

The traveler remains the only moving person. He calmly walks through the paused market, takes one apple from a wooden basket, bites it, places a crooked loaf of bread upright, lifts a coin from a merchant’s table, examines it, then returns it exactly where it was. He gently rotates a hanging wooden sign so it faces forward, walks past the motionless donkey, and smiles at the strange frozen scene.

The camera keeps moving backward in front of him, then slowly circles to his side and finishes with a close-up as he walks toward the lens, eating the apple.

Natural medieval textures, realistic stone, wood, mud and fabric, stable anatomy, realistic animals, coherent background figures, subtle film grain, no dialogue, no logos, no readable text, no floating objects, no cuts.

Original post ↗

Where it shows up alongside other effects

Phone travel vlog in Switzerland

Shot-by-shot timeline · Seedance 2.5 per the author · Oct 28, 2026

The prompt says what the footage is not (a travel commercial, a cinematic film) before any scene, then alternates selfie mode and rear-camera clips across six timed scenes. Phone behaviour is listed as specific faults: small framing corrections, autofocus changes, exposure shifts and simple cuts as if edited on a phone. The three spoken lines sit at the start, middle and end, so the speaker frames the clip.

Atmosphere and realismPerformance and emotion

Read @AIwithNatalia's prompt
Create a 30-second realistic iPhone travel vlog featuring the same girl exploring Switzerland. The entire video should feel like a casual vlog filmed by a real person on an iPhone, not a professional travel commercial or a cinematic film.

Scene 1 (0–5 seconds): A casual selfie video of the girl standing on a beautiful Swiss street with mountains in the background. She smiles naturally at the camera and says, “Hey guys! I’m finally in Switzerland, and look at this view!”

Scene 2 (5–10 seconds): Handheld iPhone footage as she walks through a charming Swiss village. Show cute houses, flowers on balconies, people walking in the background, and little spontaneous camera movements. She briefly turns the camera toward the scenery.

Scene 3 (10–15 seconds): She films a beautiful lake with mountains behind it, then switches back to selfie mode and says, “Guys, this place literally doesn’t look real!” Let a little wind and natural background noise remain in the audio.

Scene 4 (15–21 seconds): A casual café moment showing her holding a coffee, taking a bite of Swiss chocolate, and laughing softly. Include slightly imperfect framing and natural movements, like a friend casually recording her.

Scene 5 (21–26 seconds): Quick handheld clips from a scenic train ride, capturing green fields, little houses, and mountains through the window. The footage should feel spontaneous, as if she is recording whatever catches her eye.

Scene 6 (26–30 seconds): Back to selfie mode outdoors. She smiles at the camera and says, “I swear, I could stay here forever. Okay, guys, see you in the next vlog!”

Camera and realism instructions:

Shoot entirely in a natural iPhone vlog style, with handheld selfie footage and casual rear-camera clips.
Use realistic iPhone exposure, natural colors, subtle HDR, authentic skin texture, and ordinary smartphone sharpness.
Include slight hand movements, tiny framing adjustments, natural autofocus changes, and occasional exposure shifts.
Use natural daylight, realistic shadows, ambient street sounds, gentle wind, and café background noise.
Keep the girl's facial features, hairstyle, outfit, and appearance consistent across every scene.
Make her expressions, gestures, walking, and speech feel spontaneous and unscripted.
Use simple cuts between clips, like a real travel vlog edited on a phone.
Avoid dramatic cinematic lighting, drone shots, artificial slow motion, excessive color grading, perfect camera movements, beauty filters, and overly polished commercial aesthetics.

Format: Vertical 9:16, 30 seconds, realistic smartphone footage, natural conversational voice, believable lip-sync, and authentic travel-vlog energy.

Original post ↗

Four Halloween monsters fight over a potion

Shot-by-shot timeline · Seedance 2.5 per the author · Oct 26, 2026

Four characters are separated by silhouette and one signature trait each (fangs, a sculptural hat, a glowing pumpkin helmet, floating gauze), and a negative list rules out the stock versions of those costumes. Five timed shots follow a trailer structure: tease, motive in dialogue, brawl, an interruption that stops everyone, then a freeze-frame poster. Unlike most cases here it asks for a title and tagline on screen, which goes against the usual advice to add text in post.

Product, cars and trailersClose combat and impactPerformance and emotion

Lightning hero against a lake monster in a city park

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 16, 2026

Six seconds of a calm park walk set a baseline before the lake erupts, so the reveal has contrast. The hero is locked with full wardrobe and an explicit note that he never transforms, which removes one common source of drift.

Energy, spells and blastsGiants, armies and epic scaleTransformation and style shiftsPerformance and emotion

Read @CharaspowerAI's prompt
Ultra-realistic fantasy action blockbuster, 30-second cinematic sequence. Shot on ARRI Alexa 35 with anamorphic lenses, dynamic tracking, elegant orbits, aggressive push-ins, dramatic speed ramps and premium slow motion. Bright daylight in Central Park, polished Hollywood realism, realistic water spray, wind, debris and high-end lightning VFX. Spectacular, intense, non-graphic.

Main hero: one adult man, early 30s to 40s, dark short hair, light beard, black jacket, dark shirt, blue jeans, black-white-red high-top sneakers. Calm, confident, powerful presence. He never transforms. He naturally controls lightning with his hands.
Creature: one massive original aquatic monster, reptilian and powerful, huge silhouette, clear readable design.
Central Park lakeside in bright daytime, trees, benches, open lawn, pedestrians, summer light, glimpses of the skyline.

0-6s: Frontal tracking shot The man walks calmly beside the lake in Central Park. Relaxed daytime atmosphere, joggers and families in the background, light breeze in the trees.

6-11s: Massive reveal The lake suddenly erupts and a gigantic aquatic creature rises from the water, sending spray everywhere. Birds scatter, people move away fast, leaves and benches shift from the shockwave. No injuries shown.

11-18s: Power awakening The man stops and turns toward the creature. Tiny blue-white electrical arcs begin dancing between his fingers. Close-ups on his hands, shoes planting on the ground, water rippling, tree leaves lifting, static energy building in the air.

18-24s: Acrobatic confrontation The creature rushes toward him. He evades with elegant fast movement: one sharp sidestep, one low duck, one spinning reposition. Camera orbits in slow motion as sparks crackle around his hands and the air vibrates with electric tension.

24-30s: Epic lightning payoff He plants both feet, raises one arm toward the sky, and calls down one colossal blue-white lightning bolt from above. The bolt slams into the creature and the lakeside in a blinding flash, sending a huge cinematic splash and shockwave outward. The creature is pushed back into the lake and disappears beneath the water. The man stands calm in the foreground, electricity fading around his hand.

No transformation, one hero only, one creature only, premium lightning VFX, readable Central Park geography, elegant slow motion, stable anatomy, strong daylight realism, non-graphic action, no gore, no text, no logos.

Original post ↗

Salt-flat horde cleared by stored sword marks

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 11, 2026

Two invented abilities come with rules: sword strokes leave glowing marks that stay on the ground, and a dash is a visible line with a start and end. The finale detonates the earlier marks, and a finite enemy count is cleared by 27 seconds, leaving three seconds to sheathe.

Giants, armies and epic scaleAtmosphere and realismChases, racing and fast movementPerformance and emotionEnergy, spells and blastsClose combat and impact

Daylight first-person shooter trailer

Shot-by-shot timeline · Seedance 2.5 per the author · Sep 8, 2026

The daylight-only rule is stated globally and again in the beats, and the player is shown only through hands and gear. Two squadmates are separated by silhouette, and short radio lines are slotted into specific beats.

Chases, racing and fast movementPerformance and emotionAtmosphere and realism

Read @CharaspowerAI's prompt
Ultra-realistic AAA first-person shooter game trailer with premium next-gen cinematic quality. DAYLIGHT ONLY from beginning to end. Bright natural sunlight, blue sky, hard shadows, warm concrete, dust, glass reflections, heat haze and realistic outdoor contrast. Immersive first-person perspective, fast tactical movement, dynamic camera motion, polished blockbuster pacing, no gore.

Main playable character: elite adult operative seen mostly through gloved hands, sleeves and consistent gear in first person. Supporting squad: two adult teammates, clearly distinct silhouettes and equipment.

Sunlit coastal megacity, modern plazas, rooftop construction sites, bright office towers, highway interchange, open harbor, dry industrial district and elevated train lines. No night, no dark interiors.

 0-5s: Immediate daylight hook First-person view on a high rooftop under intense morning sun. Wind pushes loose fabric and dust across the roof. The player sprints forward and jumps onto a zipline between two bright glass towers. Radio: “Echo, move now.”

5-10s: Fast urban infiltration POV crashes through an open balcony doorway into a sunlit office floor, crosses the room, vaults desks, exits through another balcony and continues across connected rooftops. Strong sunlight flashes across glass and metal.

10-16s: Action escalation First-person sprint through a crowded construction zone, slide beneath a moving barrier, leap over a narrow gap, use quick tactical movements against distant opposing silhouettes. Dust bursts, concrete fragments and bright muzzle flashes stay stylized and non-graphic.

16-22s: Vehicle set-piece POV inside a fast armored vehicle racing through sunlit city streets. Palm trees, market tents and road barriers rush past. The vehicle drifts around a bus, hits a ramp and clears a broken section of road.

22-27s: Signature gameplay moment Back on foot under harsh afternoon sunlight. Player wall-runs along a construction frame, grabs a hanging cable, swings onto another rooftop and instantly deploys a compact recon drone over the harbor.

27-30s: Final daylight reveal The player reaches a huge open-air industrial yard flooded with golden late-afternoon sunlight. A massive armored machine powers on across the yard while dust rolls between them. Unknown voice over radio: “You were never sent here to win.” Player slowly raises weapon. Hard cut.

Daylight only, no night scenes, no dark blue lighting, bright readable environments, next-gen photorealistic game visuals, immersive first-person camera, stable hand and gear continuity, realistic daylight exposure, cinematic heat haze, believable movement physics, premium AAA trailer pacing, no gore, no graphic injury, no readable text, no logos.

Original post ↗

Fire fighter against a feathered demon, ink style

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 8, 2026

Ten three-second groups, each with three timed points, cut on continuing attacks so the edits match action. Heavy hits flip to monochrome ink frames without freezing time, and the duel deliberately ends undecided.

Close combat and impactEnergy, spells and blastsChases, racing and fast movementPerformance and emotion

Giant tailor scissors and red thread in a doll factory

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 7, 2026

The weapon's mechanics are spelled out (two blades on one pivot that never split) and the thread only passes through existing seams and rings, tightening gradually into a net. A fixed route around the hall and one bullet-time window keep the long take organised.

Giants, armies and epic scaleChases, racing and fast movementClose combat and impactPerformance and emotion

Formula 1 documentary trailer

Shot-by-shot timeline · Seedance 2.5 per the author · Sep 6, 2026

An emotional arc (quiet preparation, pressure, the race, silence after the finish) drives the beats instead of action peaks. Documentary camera choices and one consistent driver identity hold it together.

Chases, racing and fast movementAtmosphere and realismPerformance and emotion

Fire fist against a peacock demon, hand-drawn ink

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 6, 2026

Ten three-second blocks are for planning only, with no cuts between them, and each beat names the attack, the defence and whether it lands. Fire flares only on contact, and the arc moves from even trading to the opponent pressing to the hero winning.

Energy, spells and blastsClose combat and impactPerformance and emotionChases, racing and fast movement

Tiger cub wildlife documentary

Shot-by-shot timeline · Seedance 2.5 per the author · Sep 5, 2026

No action at all: two animal subjects, natural light and small behaviours such as a stumble and a pounce at a leaf, paced with voice-over. It shows the model handling restraint when the prompt asks for it.

Atmosphere and realismGiants, armies and epic scalePerformance and emotion

Hand-to-hand duel in a volcanic basin

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 4, 2026

Every action needs a start, path, contact point, force direction and next position, and initiative swaps every one to two seconds. The author said the model misread the intent around the three-second mark and kept the take anyway.

Close combat and impactPerformance and emotionEnergy, spells and blastsChases, racing and fast movement

Halberd heroine against a demon army at a pass

Shot-by-shot timeline · Seedance 2.5, inferred · Sep 2, 2026

The army is built in three layers (foreground threat, middle attack line, distant mass), and all effects must attach to the weapon edge or ground cracks. A no-blood rule turns hits into ash and armour fragments.

Close combat and impactGiants, armies and epic scalePerformance and emotionEnergy, spells and blasts

Bare-knuckle temple duel continuing a first frame

Continuous action chain · Seedance 2.5, inferred · Sep 1, 2026

Built to continue from a reference first frame: wounds visible at frame zero persist and accumulate, and bystanders retreat and never join. Time-ranged prose paragraphs set three to five state changes per second.

Close combat and impactEnergy, spells and blastsCamera moves and transitionsPerformance and emotion

Petal-blade swordswoman against a grassland horde

Shot-by-shot timeline · Seedance 2.5, inferred · Aug 31, 2026

A single-source rule governs the effects: every petal blade comes from the one sword and reforms into it, and the gourd only adds mist to existing petals. The horde refills locally so the field is not cleared too early.

Giants, armies and epic scaleEnergy, spells and blastsPerformance and emotionCamera moves and transitions

Giant octopus filmed from a yacht

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 29, 2026

The camera never leaves the yacht, and the clip relies on bad zoom, missed focus and off-camera shouting for realism. The tentacle rises slowly, so scale builds through restricted viewpoint rather than spectacle.

Giants, armies and epic scaleHorror, suspense and POVPerformance and emotion

Classroom prank in phone footage

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 21, 2026

Secret phone filming is described through its faults: bad autofocus, accidental zoom, suppressed laughter near the mic. Explicit guardrails (adults only, harmless, no injury) keep the prank safe.

Close combat and impactPerformance and emotion

Jungle POV with a saber-toothed cat

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 20, 2026

Only hands and a machete are visible, and sound arrives before the creature does: breathing, insects, then the threat. The reveal is staged slowly through foliage brushing the lens.

Atmosphere and realismGiants, armies and epic scalePerformance and emotion

Read @CharaspowerAI's prompt
Ultra-realistic first-person survival adventure, intense cinematic realism. Entire sequence from the point of view of an adult man moving through a dense tropical jungle. Rough handheld POV with natural body sway, heavy breathing, foliage brushing the lens, humid atmosphere, subtle lens dirt, raw practical sound, no cinematic polish. Harsh daylight filtering through wet leaves, dense mist, insects, mud, heat and oppressive jungle tension.

POV character: adult male explorer or survivor, only hands, forearms and a machete occasionally visible. Main creature: one massive saber-toothed cat, muscular, terrifying, animalistic, realistic fur, heavy feline movement, huge fangs, intense predatory stare.

Dense tropical forest with thick vines, giant roots, muddy ground, hanging leaves, shallow puddles, shafts of light, distant animal sounds and narrow overgrown paths.

0-5s: POV moving forward carefully - The man advances through the jungle, pushing aside thick leaves with one hand while holding a machete in the other. Heavy breathing close to the microphone. Insects buzz, branches crack underfoot, the path is tight and humid.

5-10s: POV slowing down - He suddenly stops. A low growl is heard somewhere ahead. The camera turns slightly left and right, searching through the dense vegetation. The machete rises higher into frame as tension builds.

10-15s: POV reveal - Through the leaves, a giant saber-toothed cat emerges into partial view, crouched low in the mud. Close view of its fur, shoulders, long fangs and fixed predatory eyes. It growls deeply and takes one slow step forward.

15-20s: POV confrontation - The man stumbles backward, raises the machete defensively, breathing faster. The creature suddenly lunges toward the camera through the foliage. The frame shakes violently, leaves hit the lens, and the shot cuts abruptly at the moment of impact.

Photorealistic jungle survival realism, immersive POV continuity, realistic feline anatomy, believable saber-toothed movement, tense practical sound, natural body motion, stable machete visibility, oppressive tropical atmosphere, no third-person shots, no fantasy cartoon style.

Original post ↗

Slap-sport championship final

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 18, 2026

Absurd content played completely straight with broadcast-style framing, close-ups and crowd reactions. Two finalists are contrasted by look and attitude so the cutaways stay clear.

Giants, armies and epic scaleAtmosphere and realismPerformance and emotionSport, dance and rhythm

Monkey banana heist through a hotel kitchen

Shot-by-shot timeline · Seedance 2.5, inferred · Aug 18, 2026

Each stage lists a starting state, main events and an ending state, so beats hand over cleanly. A deadpan side-eye from the reference image is defined as a specific head-still, eyes-move behaviour.

Chases, racing and fast movementPerformance and emotionCamera moves and transitions

Frozen time in a city park

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 16, 2026

A reference image fixes the man's identity, and the frozen world includes birds and a dog suspended mid-motion. One continuous Steadicam move ends on a hero close-up.

Atmosphere and realismPerformance and emotionSport, dance and rhythm

Read @CharaspowerAI's prompt
Ultra-realistic cinematic frozen-time drama in bright daylight. One seamless 30-second continuous shot. Smooth Steadicam tracking, subtle push-ins, one elegant orbit and a final hero close-up. Warm summer light, soft breeze, lush green park, premium live-action realism, tasteful blockbuster emotion, no cartoon look.

Main character: use (image_1) as the strict identity of the adult man, preserving his face, hair, beard, black jacket and body proportions. He wears fitted blue jeans and red, white and black high-top sneakers. Secondary character: one attractive adult woman jogging in stylish athletic wear. Background: a few park visitors, one dog, and birds in flight.
Large city park in daytime with tree-lined paths, benches, grass, soft sunlight, joggers, a dog crossing the path and birds flying overhead.

Photorealistic frozen-time illusion, stable identity, elegant body movement, coherent suspended birds and dog, realistic park atmosphere, seamless camera continuity, emotional heroic payoff, natural anatomy, no readable text, normal proportions without stretch

0-8s: Medium tracking shot moving backward - The man walks calmly toward camera through the park. Sunlight filters through the trees. A woman jogs into frame behind him from the side, moving fast along the path. Birds cross overhead and a dog runs nearby.
8-12s: Push-in to close-up - The jogging woman catches her foot and begins to lose balance, starting to fall forward. At the exact same moment, the man turns, raises one hand and snaps his fingers. Instantly everything stops. The woman freezes mid-fall, birds hang motionless in the air, the dog freezes mid-run, nearby people stop in place.
12-22s: Slow orbit around the man - He remains the only moving person. He walks calmly through the frozen park, steps between the motionless birds and people, reaches the woman, gently supports her and places her into a safe upright balanced posture as if stabilizing the moment itself.
22-30s: Low-angle dolly backward into hero close-up - He returns to his path and snaps again. Time resumes naturally. The woman regains her footing smoothly and continues without crashing, briefly confused but unharmed. Birds fly again, the dog keeps running, and the man walks forward with a subtle satisfied smile.

Original post ↗

Rooftop parking-deck MMA fight

Continuous action chain · Seedance 2.5, inferred · Aug 14, 2026

Every technique sets up the next (strikes open a takedown, the takedown draws a defence, the defence leads to the ground), and each fighter's reach shapes his tactics. Effects are limited to sweat, dust and short motion trails.

Close combat and impactEnergy, spells and blastsChases, racing and fast movementPerformance and emotion

Camcorder sighting in a misty forest

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 8, 2026

A frightened operator, focus that struggles between trees and whispered lines build to a limited but clear reveal. The creature is never played for comedy.

Horror, suspense and POVAtmosphere and realismPerformance and emotion

Read @CharaspowerAI's prompt
Ultra-realistic amateur cryptid sighting footage. Cheap handheld camcorder look, low-quality image, unstable zoom, bad autofocus, heavy digital grain, compression artifacts, shaky framing, harsh breathing close to the microphone, accidental overexposure through the trees, raw found-footage realism. No polished cinema look.

One unseen camera operator only, frightened adult hiker breathing heavily behind the camera. One massive Bigfoot-like forest creature: tall, broad, covered in dark matted hair, humanlike posture, heavy natural movement, never cartoonish.

Dense forest in daylight, uneven ground, wet leaves, tree trunks, foggy depth, broken branches, deep natural shadows.

Authentic low-grade camcorder texture, believable forest sound, realistic creature anatomy, limited but clear reveal, unstable autofocus, strong viral payoff, natural movement, terrifying realism, no comedy, no clean digital polish.

0-5s: Shaky handheld POV - The camera moves through the forest, breathing loud, branches hitting the lens. The operator whispers in French: “Attends… attends… c’est quoi ça là-bas ?” The image struggles to focus between trees.

5-10s: Messy zoom-in - Through the trees, a huge hairy figure is barely visible in the distance, half-hidden behind a trunk. The operator mutters: “Non… non… c’est pas possible…” The creature slowly steps forward into a small patch of light.

10-15s: Unstable push and panic reframing - The Bigfoot-like creature turns its head toward the camera. Clear view for only a second: massive shoulders, long arms, wet dark fur, heavy breathing. The operator gasps and whispers: “Oh mon Dieu…”

15-20s: Violent shake, partial loss of frame - The creature suddenly lets out a deep roar and takes two fast steps toward camera. The operator panics, stumbles backward, drops the camera toward the ground. The last image catches the creature rushing closer through the trees before the footage cuts to black.

Original post ↗

Lost in the woods, then the cabin

Shot-by-shot timeline · Seedance 2.5 per the author · Aug 6, 2026

A trembling flashlight POV, crying narration and escalating lines move the viewer from forest to cabin interior. Tension rises through sound and restricted light rather than visible threat.

Horror, suspense and POVPerformance and emotionAtmosphere and realism

Read @CharaspowerAI's prompt
POV found-footage horror scene in an original forest-lost-survival horror aesthetic. The entire scene is from first-person perspective, as if filmed on an old handheld consumer camera. Low-quality image, unstable autofocus, dirty digital grain, compression artifacts, low dynamic range, weak exposure, occasional glitch flicker, raw amateur camera texture.

Deep forest at night, wet ground, dead branches, distant wind, then inside a decaying abandoned wooden house hidden in the trees. Mood: extreme tension, suffocating dread, oppressive silence broken by panicked breathing and quiet sobbing.

0–5s: shaky POV moving through dense dark forest with flashlight beam trembling. Heavy breathing close to the microphone. The person whispers through tears: "I don’t know where I am… please… please..." Branches scrape the lens, footsteps crunch on wet leaves.

5–10s: the camera stumbles toward the old house. Fast unstable push through the doorway into a rotten interior. Floorboards creak. The breathing becomes more frantic. The person starts crying harder and says: "Someone is in here… I heard it…"

10–15s: slow pan across the dark room, filthy walls, broken furniture, hanging fabric moving slightly. The camera catches a childlike bundle of sticks hanging in a corner. A faint noise comes from upstairs. The person gasps, starts trembling, and whispers: "No… no… don’t do this..."

15–20s: the POV turns toward a doorway. In the far corner of the room, a motionless human figure stands facing the wall. The camera freezes, breathing turns into sobbing. The person whispers: "Why are you standing there?" The figure slowly tilts its head without turning around. Hard camera glitch, violent drop in the frame, instant blackout.

Authentic raw POV horror, realistic panic breathing, crying voice, unstable handheld motion, oppressive forest and house atmosphere, no cinematic polish, viral terrifying payoff, normal proportions without stretch.

Original post ↗

Fashion shoot in the middle of a war

Shot-by-shot timeline · Seedance 2.5, inferred · Jul 31, 2026

The clip opens on the exact composition of the reference image, then breaks into fast editorial cuts of eyes, hands and texture. Fashion posing and battle effects alternate on a set timeline.

Atmosphere and realismChases, racing and fast movementPerformance and emotionTransformation and style shifts

Read @CharaspowerAI's prompt
Use the provided image as the strict visual reference for the warrior’s face, blonde hair, armor, fur jacket, leather boots, sword, body proportions, ruined battlefield, burning fortress, smoke, embers, and dark cinematic atmosphere. Preserve her identity and outfit throughout the entire sequence.

Create a 20-second impossible high-fashion warrior photoshoot, blending sensual confidence, brutal fantasy action, luxury editorial posing, explosive VFX, and aggressive cinematic camera work. The result must feel like a viral fashion campaign filmed in the middle of a medieval war.

0–4 seconds — Iconic opening pose
Begin with the exact composition of the reference image. The warrior stands on shattered stone, sword lowered beside her, staring directly into the lens. Her hair and fur move violently in the wind while embers spiral around her. The camera rushes toward her with a low-angle dolly-in. A powerful white flash freezes her pose, followed by a horizontal blue lens flare crossing her eyes and armor.

4–8 seconds — Fashion meets combat
She walks toward the camera with controlled, sensual confidence as explosions erupt behind her. Use rapid editorial cuts: close-up of her eyes, lips, hand tightening around the sword, fur texture, armored waist, boots crushing burning debris. She suddenly pivots and blocks an incoming blade from an unseen attacker. Sparks explode across the lens as the camera whip-pans around her.

8–12 seconds — Impossible action photoshoot
She launches into a fast spinning sword movement, cutting through smoke and shattered banners. Freeze her mid-motion in a dramatic fashion pose as dozens of camera flashes fire from impossible angles. Add elegant motion echoes, metallic silhouette trails, floating reflective shards, streaking embers, and anamorphic lens flares. Each flash captures a different powerful pose before snapping back into live action.

12–16 seconds — Sensual hero moment
The action suddenly slows. She plants the sword into the ground, steps close to the lens, and leans slightly forward with a fierce, provocative gaze. Wind sweeps her hair across her face while warm firelight shapes her silhouette. The camera performs a tight orbit around her body and rises into a close-up. A strong golden lens flare blooms behind her shoulder as ash drifts through the frame.

16–20 seconds — Epic viral finale
A massive explosion destroys part of the fortress behind her. Without looking back, she pulls the sword free and turns sharply toward the camera. Use an extreme speed ramp as she swings the blade directly past the lens, creating a seamless transition into three rapid flash-frozen editorial poses: sword over her shoulder, blade pointed downward, then a dominant frontal stance.

Finish on a slow-motion hero frame with the warrior standing perfectly centered, smoke and fire behind her, embers suspended in the air, her hair flowing, armor glowing along the edges, and a huge cinematic lens flare crossing the screen.

Ultra-premium fantasy fashion editorial, sensual but powerful warrior energy, impossible battlefield photoshoot, luxury magazine attitude, brutal sword choreography, low-angle camera, macro beauty shots, whip-pans, fast orbit movement, snap zooms, speed ramps, flash photography, freeze frames, motion echoes, metallic VFX, realistic fire and debris, dramatic anamorphic lens flares, realistic hair and fabric physics, no dialogue, no text, no subtitles, no identity drift, no wardrobe change, no facial deformation, no extra limbs.

Original post ↗

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

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