NewOpenClips MCP is live.Your AI video studio, inside any AI chat.See how it works →
Pricing

Action & movie

Fights, chases and stunts. Starring you.

Upload one photo. Pick a scene. Get the take with you in it.

Inspired by @SirGlavan_on X

Four steps, one generation each. Step 1 pick any avatar (@avatar). Step 2 turn the visitor's product photo (@product) into a three-view product map on grey so every later step sees the same details. Step 3 a two-frame reference sheet that locks the face, the hair, the jacket and the chained product together (Nano Banana Pro holds the face). Step 4 one unbroken 30s 16:9 Seedance 2.5 render, reference mode, with the sheet and the product map attached. The prompts never name the product or a gender, they read both from the images, so the recipe works with any avatar and any product the courier can hold in their hands. Known traps: never write 'a held beat of silence', the model renders it as a literal frozen frame, write tension as micro-motion instead; never write 'walks to the doors', the model walks to the END of the car, so stage the fight to finish beside the side doors; never cut to an outside camera on a door that already opened, the model closes and reopens it.

4 steps6854.1K
Seedance 2.530s · 16:9
Inspired by j_smeaton99on X

30-second 16:9 ultra-realistic AAA third-person action gameplay. Sharp Unreal-like capture, not a cinematic montage. @image1 is absolute character truth (face and sci-fi suit). @image2 is the environment. Photocopy both. Do not describe appearance. Use they / their / them only. SCI-FI world, not fantasy Korea. One colossal chrome-white mech tiger guardian. 0–4s: Sprint up damaged metal stairs toward the fortress. Jump a broken span, grab a ledge, climb into a ruined courtyard. 4–8s: They pick up a glowing pale-blue relic in a shrine-like console. Structures and glyphs light. 8–12s: Ground cracks. A colossal mech-tiger guardian rises. Scale read. 12–17s: Guardian wrecks the gantry. They dodge falling beams, slide, jump a broken energy bridge, swing a cable over the ravine. 17–21s: Relic dash — fast blue trail across broken platforms. 21–25s: Guardian chase. Dodge-roll a energy strike. 25–28s: Short blade, strike the glowing weak point on the shoulder. Blue pulse. 28–30s: Guardian roars, they take a defensive third-person stance. Boss fight about to start. No on-screen text, no HUD, no logos. AUDIO: boots, metal, relic hum, roar, no lyrics.

4 steps13410.4K
Seedance 2.530s · 16:9
Inspired by umesh_aion X

Hyper-realistic 15-second 16:9 continuous ice action. Sound ON. OPENING: Frame 1 is already ON THE ICE, mid-stride. Never a studio portrait or grey still. The person is exactly the attached portrait. Blue-white-red vertical skating suit, white blades. No other people. No on-screen text. PHYSICS — MANDATORY (this failed last time): Sea ice is heavy, thick, slow. Plates do not teleport, stretch, or explode into CGI shards for no reason. A crack opens from a real stress line and widens at a believable speed. When they step on a floe it tilts under their WEIGHT, then they recover. Jumps need a visible crouch, takeoff, flight, and a heavy landing that shakes the plate. No sliding through solid ice. No ice appearing from nowhere. Gravity always down. Camera can move, but the world stays physically consistent. 0–5s: They skate a flat sheet. ONE crack races under the blades. They hop it. The ridge behind them lifts slowly, not a firework. 5–10s: The sheet splits into TWO or THREE large floes. They commit to one plate; it rotates slowly. They skate across it. 10–15s: They jump a dark-water gap onto the next floe, land hard, then a last push onto the snowy bank. Ice behind them is a few big tilting plates — not confetti. AUDIO: blades, ice groan, crack, splash, wind. No voiceover. Duration 15s. Aspect 16:9.

2 steps627.3K
MiniMax H315s · 16:9
Inspired by @azed_aion X

Create a 15-second photorealistic sci-fi action sequence in five shots: a sprint along a collapsing cliffside trail, a deliberate leap into the canyon, freefall through golden light, a mid-air catch by a rising open-cockpit skimmer, and a climb above the clouds into sunset light. Golden-hour backlighting, volumetric clouds, realistic flight physics and cloth simulation, premium blockbuster VFX. Your avatar is the rider in every shot.

2 steps845.2K
Seedance 2.015s · 16:9
Inspired by lansenaion X

15-second 16:9 stylized game-cinematic. Painterly oil-clay surfaces, chunky sculpted cloth, grim ruined street, high-contrast acid-green magic, heroic third-person camera, stop-motion clay texture. Not photoreal. @image1 is absolute character truth. Photocopy the sheet. Do not describe appearance, hair, clothes, or face. Use they / their / them only. 0–3s: Extreme close-up. A palm gathers a dense ball of acid-green fire. Hat brim frames the shot. Slow raise of the hand. 3–6s: Pull wide. They stand in a glowing green rune circle on a ruined street at dusk. Monster silhouettes pack the wreckage ahead. Hand still rising. 6–9s: Green flame detonates from the palm. Shockwave. A tide of green-veined creatures fills the ruins. 9–12s: They lift off the ground. Floating beams smash through the horde. The whole street is inside the cast. 12–15s: Snap to a close grin, eyes catching the green light. Camera holds as the street floods with the spell. AUDIO: spell charge, explosion, creature roar, no lyrics. No on-screen text.

3 steps1213.8K
MiniMax H315s · 16:9
Inspired by @Iancu_aion X

Generate a cinematic video of a fighter, fighting the swarm of enemies.

4 steps502.1K
Seedance 2.530s · 1:1
Inspired by @Artedeingenioon X

Generate 15-second anime music video (16:9, 2K). Pure 2D cel animation + torn-paper collage aesthetic (ink black, warning red, paper white). Female swordfighter protagonist with antagonist silhouette shadow-pursuer. 8 cinematic shots with hard cuts. Kinetic typography synchronized to action. Original K-pop hip-hop × cyber-grunge soundtrack with foley. Character lock: protagonist identical throughout. Antagonist appears only in shots 3, 6, 8. Two still moments for pacing. Hard cut to black on final beat.

3 steps181.6K
MiniMax H315s · 16:9
Inspired by @techhallaon X

CHARACTER LOCK - HIGHEST PRIORITY: The person in every frame is exactly the person in @avatar. Preserve their facial identity, facial structure, eye shape and spacing, eyebrows, nose, mouth, exact skin tone, hairline, hairstyle, hair length, texture and colour, freckles and distinctive marks, apparent age, build and body proportions - identically, shot to shot. Preserve their gender presentation exactly as shown; do not masculinize, feminize or otherwise reinterpret it. Do not beautify, slim, de-age or restyle them. No identity drift. THE HEAD IS ALWAYS BARE: no hat, no cap, no hood, no sunglasses and nothing covering the hair, the eyes or the face at any point. The face and hairline stay visible in every shot, including the leap, the fight and the climb. ATTACHED IMAGES ARE MANDATORY AND FINAL: If an avatar image is attached, that is the person - reproduce them exactly in every frame. The attached images outrank every word below them; wherever image and text disagree, follow the image. @avatar defines identity only - its clothing, pose, background and lighting are ignored completely. Wardrobe (comes from this text, never from @avatar): an unbuttoned cream linen shirt with the sleeves rolled, worn loose and open over a plain white cotton tank, sand-coloured linen drawstring shorts to mid-thigh, a thin woven bracelet and worn canvas sneakers. The shirt hangs loose and is never fitted to the chest. Unisex cut, no logos or lettering. Fit each garment to the character's body naturally without altering its design, cut, proportions or colour - the garment is fixed, only its drape adapts. The same in every shot. Ultra-realistic cinematic action cinematography, 30 seconds. @avatar is hunted across a modern Greek island town on a holiday afternoon. Hard Mediterranean sun, white and bone against Aegean blue, handheld weight, real motion blur, shallow depth of field, fine grain. Hands stay correct in every frame: five fingers per hand, natural grip on every ledge and wrist. 0-4 seconds - THE LEAP. The film opens already in the air: @avatar mid-leap off a whitewashed rooftop terrace, arms spread, the open shirt snapping, falling toward the lower roofs. No fade in and no still opening frame. Below spread white cube houses, blue domes, stepped lanes and the flat blue sea. The camera falls alongside. 4-8 seconds - THE LANDING. @avatar drops onto a lower terrace into folded fishing nets, rolls out in one motion and comes up running along the white roof. The camera whips down and picks up the run from behind at shoulder height. 8-12 seconds - THE LANES. @avatar sprints a narrow stepped lane between white walls, ducking a laundry line and cutting between tavern tables while tourists scatter. Bougainvillea petals and dust cross the lens. 12-17 seconds - THE FIGHT. Two hired killers in dark linen suits step out of a doorway and reach in. @avatar turns the first one's arm aside and puts them both down on the steps in two fast moves. The camera comes in close on @avatar's face as the arm turns and holds there, the face fully readable. The killers are unarmed in frame: no gun and no knife is ever drawn, and there is no blood and no gore. 17-22 seconds - THE CHASE. @avatar runs a shadowed alley with two more suited killers closing in behind, faces turned away, vaults a terrace wall, slides under a stone arch and takes a corner hard past a blue-domed church. The camera stays close behind and unsteady. 22-27 seconds - THE CLIMB. @avatar hits a white wall of window ledges, a pergola beam and a drainpipe and goes straight up it, hand over hand, back onto the roofs. The camera rises and turns to show the town dropping away. 27-30 seconds - THE TERRACE. @avatar pulls onto the highest terrace, rises out of a crouch and turns to look straight into the lens, face fully readable. Behind: white roofs, blue domes, the harbour and the whole sea under a low sun. Final wide, held. The killers and the tourists are anonymous - hats, turned backs and faces out of frame. Nobody else wears sunglasses and no other face is ever the subject of a shot. Sound: breathing and footfalls all the way through, linen snap in the fall, nets and dust on the landing, tavern voices and cicadas, scuffs in the fight, sneakers on stone steps, wind and gulls on the last terrace, under a restrained low score. No subtitles.

2 steps131.2K
Seedance 2.530s · 16:9
Inspired by itsSaira_1on X

@image1 is CATE — photocopy her exact face, bone structure, hazel-green eyes, long wavy PASTEL-PINK hair. Do not dye her blonde. IGNORE any graphic tee in @image1. WARDROBE FROM THE SCENE: white off-the-shoulder sweater, black pleated mini-skirt, backpack. She MOVES. Cinematic photoreal urban daylight. Glowing RED eyes VFX, PINK energy trails, super-speed motion blur, slow-motion impacts. Cops, masked criminals, hostage in beige coat, police car are BEATS. CATE is the hero. 30-second 16:9. No on-screen titles. No logos. [TIMELINE — 30 SECONDS] 0–2s: Outside a modern building. Three masked criminals hold a woman in a beige coat, gun aimed. Cops and a police car behind. 2–4s: Close-up CATE, white sweater, black mini, backpack. Deadpan. Eyes ignite glowing red. 4–9s: She dashes — pink trails and sparks at her feet. Punches a criminal. Gun flies in slow-mo. Hostage shocked. 9–24s: Martial arts. High roundhouse into a chest — pink lightning shockwave, floating petals. Super-speed kicks take down the rest. 24–27s: She kneels by the unconscious hostage, pink energy on her arm, then bridal-carries her to stunned officers. 27–30s: Wide street. Cops over unconscious criminals. CATE zips out — fading pink lightning, gust of wind. Audio: city, impacts, whoosh, no lyrics.

2 steps27123.8K
Seedance 2.530s · 16:9
Inspired by Arvin010717on X

Generate a video of 15s 16:9: desert, one-finger array, your avatar, fighting monsters.

2 steps1298.8K
Seedance 2.515s · 16:9
Inspired by @TanLuAIon X

Most action clips fall apart at the moment of impact. This one lives there. A hotel balcony, second floor, rain coming down hard. A hidden shooter sights a heavy mechanical crossbow through the railings and picks a target on the street below. The bolt goes out — and then it simply stops. Half an inch from a face, hanging in the air, water still running off the umbrella beside it. Nothing glows. Nothing flashes. The camera just holds there long enough for you to understand that whoever is standing under that umbrella was never in any danger. Then two fingers move, and the bolt goes home faster than it came. Fifteen seconds, five shots: the over-the-shoulder ambush POV, the ballistic descent, the frozen bolt at the profile, the finger close-up, and the low-angle hero shot in the rain. All of it from one avatar reference and one prompt — no compositing, no effect layers, no on-screen text. Every bit of the impact is camera grammar: a locked high angle through the railings, a whip pan and tilt up, one hard cut, and an 85mm lens wide open in a downpour, graded cold-iron so the single red coat is the only colour in the frame. Drop in your own avatar, any gender, and the film rebuilds around it. The deep-red raincoat is written into the prompt rather than pinned to a reference, so one edited line changes the whole look — black for straight noir, bone white against the wet asphalt, anything you like. Fifteen seconds. One render. Zero effects.

2 steps426.9K
MiniMax H315s · 16:9
Inspired by @abxxaion X

A woman serves to open a beach volleyball rally against her opponent, and the camera locks onto the ball for the entire exchange. After her serve and toss, the shot becomes one continuous take: the camera whips rapidly left and right between the two players as they trade a dig, a set, a mistimed spike and a final hard-driven kill shot into the sand — fast, breathless, and shot like it was caught live courtside rather than staged.

1 step405K
Gemini 2.5 Flash Image16:9
Inspired by @tokyo_Valentineon X

Generate a video with your animated avatar, in japanese anime world, fighting the siege.

4 steps443.8K
Seedance 2.530s · 16:9
Inspired by nastassiavideoon X

Exactly 15 seconds, 16:9. ONE CONTINUOUS UNBROKEN TAKE. Live-action fantasy comedy, 35mm grain, bright noon daylight, saturated but natural skin. Same world palette as before: floating grass islands, red-and-white mushrooms, green pipes, gold coins, waterfalls, pink-roofed castle, red-and-white curb, blue sky with big clouds. @image1 is GREEN — lock this exact face 100%. Dark bob, green cap, denim overalls, cream top, white gloves, brown boots. Green kart. @image2 is RED — lock this exact face 100%. Light brown hair, red cap, denim overalls, cream top, white gloves, brown boots. Red kart. Do not swap faces. Do not open on a studio portrait; frame 1 is already racing. Prince at the finish line, slightly to one side of the checkered stripe: blond curls, gold crown, pink costume, arms open. Directly BEHIND him, not beside: Bront the teal-scale copper-shell horned creature, white crest, brown eyes. Bront is clearly behind the prince. 0-2s — Already racing, side by side. RED shouts, exact lip sync: "My prince!" GREEN: "No! MY prince!" 2-6s — GREEN steers hard into RED and SHOVES her completely off the track. RED's red kart leaves the road entirely — over the curb, off the floating island, falling away. GREEN stays on the racing line and keeps going. This is a full knockout, not a bump. 6-10s — GREEN alone, full speed, drifts the last bend, crosses the finish line still in the green kart. Prince is ahead with arms open. She does NOT stop in front of him. She drives past him, around his left side, and brakes behind him next to Bront. 10-13s — GREEN jumps out, ignores the prince, runs around him and throws her arms around Bront. Tight hug. Bront grins. Prince turns, confused. 13-15s — Prince drops to his knees and sobs, kicking. GREEN keeps hugging Bront. RED is gone off-track. Abrupt end. One uninterrupted shot. No cuts. Audio: engines, a hard kart impact, wind, two spoken lines, hug, prince sobbing. No music, no subtitles, no logos. No on-screen text. Duration: exactly 15 seconds. Aspect ratio: 16:9.

2 steps472K
Seedance 2.515s · 16:9
Inspired by liluocheng13on X

TWO LEADS — PHOTOCOPY BOTH STILLS. Do not invent faces. Do not swap them. @image1 = FEMALE LEAD. Photocopy her face, bone structure, skin, long blonde hair, green eyes, body. ALSO photocopy her outfit from that photo (the prompt has no wardrobe for her): burgundy corset bodice with black leather harness straps and silver rings, high-slit burgundy skirt, ivory chiffon cape that ombrés to blood-red at the hems, black gladiator lace-up sandals, small hoop earring. She fights with a jian / straight sword. Same woman and same clothes every frame. @image2 = MALE LEAD. Photocopy his face, bone structure, skin, wavy dark-brown hair to the jaw, body. ALSO photocopy his outfit from that photo: wide black kasa/sugegasa straw hat, black kimono and hakama, black sash, katana at the left hip, black tabi. Same man and same clothes every frame. The hat can tilt in motion but stays on. Never give him her burgundy dress. Never give her his kasa. Never a third human. Never OPENCLIPS.AI type. SETTING: vast ornate temple hall from the source film — gilded dragon columns, wet reflective stone floor, thousands of candles, teal mist, high vault. 16:9. 30 seconds. One continuous action sequence. No dialogue, no subtitles, no on-screen text. LIGHTNING (required on every clash): whenever the two swords collide, a bright white-blue lightning arc flashes along both blades and a fork of electricity cracks the air at the contact point — not ordinary orange sparks alone. Every bind, parry and strike that meets steel = lightning. CORE: zero-delay chaining; scene changes only via their bodies (slam a column, leap from a pillar, crash through candelabra); real weight — blocked hits shove the other back, feet slip, bodies twist. Camera follows limbs; never orbit a frozen pose. BEATS (30s) 0–2s already sprinting from opposite ends of the hall, full-speed clash at centre. Blades meet with a lightning burst; both recoil half a step; floor tiles crack. No establishing pause. 2–7s close melee: he sweeps a wide horizontal cut; she drops almost to the floor, blade skimming her nose, a lock of blonde hair cut; she springs up with a straight thrust to his chest; he spins, wrists reverse to bind. Swords lock, shoulder to shoulder, then both shove off with lightning at the bind. Two steps back. 7–9s the shove sends him into a dragon column; he rebounds, runs vertically up the pillar; she explosive-jumps after. Two or three lightning clashes on the column, blades scarring the gilt. 9–12s he is forced to the beam; her heavy blow knocks him off; he mid-air flips, diving slash; she drops from the column to meet it. Landing clash — lightning + sparks, arms visibly shake. Now they are on the hall floor again. 12–17s densest round: she three accelerating cuts (he sidesteps, then floor-dodge, then ducks under and slashes at her waist). Her cape corner splits. He lunges; she barely slips, fabric tears at the shoulder; she counters so hard his blade rings and the tsuba cracks; he staggers two steps. 17–21s bind at close range, simultaneous break, charge again — they smash a row of candelabra; candles snuff then relight from the sword-wind; they keep fighting through fallen candles and oil, lightning every contact, water-mist from the wet floor. 21–26s peak speed: cut, thrust, parry, counter in two–three seconds, each hit a lightning flash plus body feedback. Teal-green sword qi spirals. No frozen frame. 26–29s simultaneous all-out strike; blades meet; a hall-wide teal-white lightning shockwave hurls both back several metres; tiles and columns crack; candles gutter; camera shake then soft defocus. 29–30s smoke thins. Both still on their feet, one knee or standing, chests heaving, sword-hands shaking from exhaustion. Remaining candles relight. Wide two-shot. Hold. No titles. AUDIO: steel, thunder-crack on every clash, stone, fire, breath, shockwave. No BGM. No speech.

3 steps531.7K
Seedance 2.530s · 16:9
Inspired by @meAsifAion X

A relaxed boardwalk stroll turns into a slick, deadpan action sequence when a bathrobe-clad parent takes down three cocky skateboarders for ruining their kid's ice cream — capped with an unimpressed freeze-frame.

2 steps151.3K
Seedance 2.516:9
Inspired by GlitterPixelyon X

Generate a high-speed dark-fantasy battle. Cinematic manga aesthetic, hand-drawn ink animation.

5 steps34215.1K
Seedance 2.015s · 54:31
Inspired by @im_shahid7on X

PLAYER 1: A player — the person from @image_1 — wearing a deep red and white jacket with beige cargo pants. PLAYER 2: A player — the person from @image_2 — wearing an oversized white and teal jacket with loose cargo pants. THE COURT: An empty urban street basketball court. Style & Atmosphere A dynamic presentation in a semi-flat editorial illustration style — bold, heavy brushstrokes, angular geometric rendering, blocks of color, and desaturated neutral concrete tones accented with rich deep red and teal. Warm sunlight shines diagonally from the left, creating clear light-and-shadow contrast with visible, delicate brush textures. Both characters are rendered fully in this illustration style, exactly as they appear in their reference images. Narrative Overview PLAYER 1 challenges PLAYER 2 on the empty court, leaves them behind with a crossover move, breaks through, dunks over PLAYER 2, and then shows off while their opponent is on their knees. Dynamic Description A steady push-in shot, medium close-up — PLAYER 1 takes large strides across the cracked asphalt from the left side of the frame and aggressively throws the basketball into PLAYER 2's chest. PLAYER 2, standing on the right beside the wrapped basketball hoop structure, catches the ball against their chest. Hard cut to a low-angle, horizontal stabilized tracking shot, medium shot — PLAYER 2 dribbles low while PLAYER 1 mirrors them closely with intense defensive pressure. The two slide from left to right along the faded free-throw arc. Sneakers screech against the ground as dust rises. A whip-pan follows the basketball into an intimate handheld shot — PLAYER 1 performs a sharp crossover, rapidly switching the ball right, left, and right again. PLAYER 2's lead foot slips and they stumble for half a beat. PLAYER 1 tears past their side and drives toward the basket. Handheld tracking shot, medium shot — PLAYER 1 gathers the ball while moving and explosively drives from left to right toward the basket on the right. PLAYER 2 is half a step behind, lunging to defend, while motion trails emphasize their speed. Hard cut to a three-quarter side stabilized tracking shot that cranes upward — PLAYER 1 jumps into the air and violently slams the ball through the damaged net. The warm low-angle sunlight bursts into a glowing halo behind the backboard. The rim shakes from the impact. The camera always remains beside the basket and is never positioned directly underneath it. Quick cut to a locked low-angle shot — the basketball drops onto the cracked asphalt and bounces once. PLAYER 2 is on both knees at the base of the wrapped steel pole, head lowered and shoulders slumped. Static Description An empty urban street basketball court with a faded free-throw area painted onto cracked asphalt. On the right stands an old basketball hoop with a damaged net attached to a wrapped steel pole. Tall concrete walls, chain-link fencing, urban apartment buildings, and rooftop water tanks fill the background beneath a bright blue sky. Warm sunlight from the left casts shadows toward the right. Audio PLAYER 1 says: "Show me what you got." Sound Effects The sound of a basketball hitting a chest, sharp sneaker squeaks, and distant low urban ambience.

5 steps977.5K
Seedance 2.515s · 16:9
Inspired by @TechieBySAon X

PAINTED ACTION SHORT. ONE PERFORMER AGAINST A PATROL. NO DIALOGUE. FIFTEEN SECONDS. ASSET MAP - every reference is bound here and cited by tag below: @avatar_1 - the tracker, and the only named performer in the film. Face, hair, skin, age, build and every physical characteristic come entirely from @avatar_1 and never from this text. IDENTITY LOCK - HIGHEST PRIORITY: the tracker is exactly the person in @avatar_1. Keep their face, hair, skin tone and build the same from the first frame to the last. Do not restyle them, do not age them up or down, do not slim or bulk the build, and never put a second or third performer in frame with a speaking part. One performer, the same one, for the whole film. DRAWN, NOT PASTED: @avatar_1 is drawn into the painted style of the film - brush texture, ink shadow, flat graphic light - and never composited in as a photograph. Their face still has to read as unmistakably theirs: the same features, the same proportions, the same hair and skin tone, translated into paint. ATTACHED REFERENCES ARE MANDATORY AND FINAL: if a reference is attached, that is the performer - reproduce them exactly, in every shot, and never invent a different one. The attached reference outranks every word written below it. FACE TIME: @avatar_1's face is clearly visible and clearly lit in the first shot and again in the last. Never bury it in shadow, never cover it, and never hold the camera behind them for a whole shot. STYLE: hand-painted animated action in the manner of a printed comic brought to motion. Thick visible brush texture on every surface, hard graphic shadows, saturated jungle greens against warm skin tones, shafts of hard sunlight cutting through canopy haze, dust and spray catching the light. Bold clean shapes, strong silhouettes, cel-animation timing with snappy holds on the impacts. Not photoreal at any point. WARDROBE LOCK: @avatar_1 wears a mud-streaked dark green oilskin coat open over a grey undershirt, a long strip of cloth wound around the left forearm from wrist to elbow, heavy laced boots, and a plain machete in a worn leather sheath at the hip. No hat, no headband, no bow, no dog tags, no insignia, no lettering on any garment. These garments are fixed for the whole film and only their drape adapts to the body wearing them. THE PATROL: four armed figures in plain olive field uniforms with soft caps, carrying rifles. Faces kept in shadow or turned away. No badges, no flags, no unit markings, no readable lettering anywhere on them. THE PLACE: deep rainforest. Buttressed roots, hanging vines, a floor of wet leaf litter, a shallow brown river running through the middle of it. Mist hangs between the trunks and the light comes down in hard vertical columns. 00:00-00:03 Close on @avatar_1 turned to camera, canopy light falling across the face, eyes up and steady, drawing the machete in one unhurried motion. The face is large in frame and fully readable. Far behind them down the corridor of trees, one of the patrol is a small shape in the haze. The camera holds. 00:03-00:06 Cut to boots at ground level, running hard through leaf litter, water and leaves kicking up past the lens. The camera tracks alongside at speed. 00:06-00:09 Cut to a medium shot in a clearing: @avatar_1 turns hard into two of the patrol at close quarters and strikes the nearest rifle aside. Fast, heavy, unglamorous movement, and the face catches the light through the turn. The camera whips with the turn. 00:09-00:12 Cut to a wide shot across the shallow river: two of the patrol stand in the water at the far bank with rifles raised, and @avatar_1 swings low and hard, sweeping the legs from under the nearest one so they go down into the brown water in a sheet of spray. The camera pushes in with the movement. 00:12-00:15 Cut to @avatar_1 alone in the shallows, coat heavy with water, the patrol down around them, a small fire burning in the wreck of a camp behind. The camera pushes in slowly until the face fills the frame, lit and clear, and holds there to the last frame. This is the first and only fire in the film - nothing burns before this shot. AUDIO: rainforest under everything - insects, distant birds, water running over stones. Boots in wet leaves, cloth and leather under strain, the dull heavy impacts of the fight, a body hitting water, fire crackling at the end. A spare percussive score built on low drums and one held low string note, kept well under the effects and never rising to a peak. No voices, no shouting, no dialogue, no lyrics. Do not generate subtitles, captions, on-screen text, a title card, a logo, a wordmark, a watermark or any readable lettering or numerals anywhere in frame. Do not generate spoken dialogue or narration.

2 steps1755.3K
Seedance 2.515s · 1:1
Inspired by TanLuAIon X

@image1 is ELARA — photocopy her exact FACE, bone structure, brown eyes. IGNORE white dress in @image1. Default hair is silver-grey waves; hair MAY change with each universe look. She MOVES. Never a frozen JPEG. 18-second 16:9 photoreal. Taiwanese street arcade then rooftop. Rapid Everything-Everywhere costume jumps on THE SAME kick. Same extra people (tank-top fighter, onlookers) stay. No on-screen text. No logos. LOOKS (only Elara changes — pose, camera, extras stay): 1 schoolgirl: white short-sleeve shirt, navy pleated skirt 2 fedora detective: black suit, fedora, boots 3 mecha: silver-purple hard-suit, short hair 4 black bladesman 5 alien 6 general 7 ninja 8 pink-girl Keep kicking leg in the foreground, rear leg bent. [TIMELINE — 18 SECONDS] 0–2s: Covered arcade, scooters, metal shutters. ELARA in schoolgirl look runs at a tattooed man in black tank + track pants. Two onlookers. 2–5s: Flying side-kick. On impact, her LOOK snaps every ~0.4s through detective → mecha → ninja → bladesman → pink-girl. Same kick. Gun/crowd freeze. 5–8s: Kick lands. He stumbles. She lands, still snapping looks. Extras unchanged. 8–11s: HARD CUT dusk rooftop. ELARA with huge red-white wings leaps over three teal-hooded swordsmen. Tokyo tower far. 11–14s: Close-up ELARA (same face) galaxy-purple wings, leather bomber, giant sci-fi energy cannon, firing. Smile. 14–18s: Wide: she flies off a rooftop explosion, cherry tree, Tokyo tower in haze. Wings beat. Designed hold. Audio: kick whoosh, impacts, crowd, wing rush, boom. No lyrics.

2 steps464.8K
Seedance 2.518s · 16:9
Inspired by @Goodmanprotocolon X

Create an ultra-realistic, cinematic fast-paced action movie scene inside the cockpit of an older twin-engine regional passenger airliner at night, flying through a thunderstorm, using intense handheld cinematography, realistic movement, dramatic sound design, and believable aircraft motion. CONTINUITY: @avatar wears the same plain dark navy long-sleeved top in every shot. The cockpit is an older analog flight deck - rows of round mechanical dials, toggle switches and worn levers, no digital screens. The only light is the green-amber glow of the instrument panel plus sudden white flashes of lightning through the windshield, with heavy rain streaking across the glass throughout. No airline branding, no logos, no readable text anywhere on the aircraft, the instruments or the clothing. 0-6s: @avatar steps into the dark cockpit and suddenly realizes both pilot seats are empty. @avatar stares at the empty seats in shock, face lit from below by the instrument glow, a lightning flash briefly filling the windshield. Use tight handheld shots, quick cuts, and subtle camera shake to create immediate tension. 6-13s: @avatar races toward the left seat and takes control. Cut between @avatar's determined expression, their hands closing around the worn control yoke, and analog needles and warning lights jumping across the panel around them. 13-21s: The aircraft slams into violent storm turbulence and jolts hard. @avatar grips the yoke and struggles to keep the plane stable while alarms, rattling metal and hammering rain fill the scene. Lightning strobes through the windshield in bursts. The camera moves dynamically with the action. 21-30s: Start on an extreme close-up of @avatar's focused expression seen from outside, through the rain-covered windshield. The camera pulls steadily backward through the glass, revealing the lit cockpit, then the nose, then the whole aircraft - a twin-engine regional passenger airliner with a plain unbranded fuselage - until it is small against enormous black storm clouds with lightning firing deep inside them. End on that wide aerial. Style: Ultra-realistic Hollywood action-film cinematography, natural textures, realistic facial expressions, handheld camera, cinematic motion blur, dramatic lighting, immersive sound effects, tense orchestral score, realistic aircraft physics, seamless editing, and strong visual continuity.

2 steps762.6K
Seedance 2.530s · 16:9
Inspired by itsSaira_1on X

Generate a video of your avatar fighting the evil ghosts.

2 steps1122K
Seedance 2.515s · 16:9
Inspired by @ou_zhen599on X

For the target video, the first frame is fully referenced from @image3 (storyboard, panel 1). integrated_multimodal_description: A cinematic photoreal adventure-thriller image-to-video sequence. @image1 (the character sheet) is an absolute character lock: face, body, hair, costume, materials and colours are taken from it exactly, with no drift or reinterpretation in any beat. @image2 (the location plate) is an absolute environment lock: the bridge, the wet stone ruins at the edge, the broken rock lip and the drop below are taken from it exactly and the sequence never leaves that space. @image3 (the storyboard) is followed exactly, beat for beat, matching its shot order, framing and pose. Preserve the exact same single character throughout: same face, bone structure, hair, skin tone, clothing, body type, age impression and humid sweat texture. Only one character is allowed. No other people, no monsters, no new scene, no interior cut, no mirrored duplicate, no extra limbs. The critical location-state chain is locked: first they are still on the bridge deck moving forward; then the bridge breaks and they fall and catch the side rope; then they remain suspended in open air with no stable foothold, and never return to the deck or the edge; only in the final beat do they catch the broken lip and climb up, for the first and only time. Beat 1 — begin on the bridge deck in a low intimate near shot. Start on their hand crushing the thick rope, then lift into their forward-leaning shoulder-and-back effort. They are still on the bridge deck, moving carefully forward. One hand presses the rope, the other steadies it, the leading foot tests forward, knees slightly bent, core tight to stabilise the bridge. Their first emotional layer is focused bridge control with panic suppressed: brows lightly drawn together, eyes fixed on the next foothold, lids tense, nostrils opened by restrained breath, lips slightly parted, jaw locked. They say quietly, <d>[English] Easy... keep moving.</d> Filtered daylight rakes the rope line and the working shoulder; clothing clings and creases under reach and rope pressure. Beat 2 — still on the bridge deck, in a frontal retreating medium shot, moving along the bridge axis toward camera. They accelerate by two quick steps. The wood planks start loosening under them and the bridge suddenly sags. They do not stop; they instantly drop their centre of gravity and keep moving. Their second emotional layer is tightening alertness: brows contract more, gaze narrows sharply, eyes flick between the planks below and the broken lip ahead, nostrils open wider, lips flatten, jaw hardens. Tiny wood splinters and droplets fall through the plank gaps, coarse rope fibres begin fraying, and the bridge shivers with elastic vibration. Beat 3 — the bridge breaks. Begin on a tight structural view of planks and rope, then follow the fall briefly. The planks snap apart, the main rope fibres burst one after another, and the bridge section folds and drops. They fall with it and instinctively grab a side rope with both hands; the rope catches them violently in midair. Make it unmistakable that from this moment they are hanging outside the edge in open air with both feet off the ground and no support beneath them. Their third emotional layer is uncontrolled terror forced into survival: pupils contract hard, eyes open wide but not caricature-wide, brow centre drives down, nostrils strain, teeth clamp, throat loses a breath on the sudden stop. They say, <d>[English] No— hold... hold it!</d> The first half breaks on the arrest; the second is forced self-stabilisation. Hard physics must read: rope fibres snap with different thicknesses and recoil, broken planks and wet wood dust tumble downward, their palms scrape the rope and shed fibres, shoulder joints, arms and back show real tension from the violent stop, hair lifts with fall inertia then drops back heavy against face and neck, straps yank tight and rebound. Beat 4 — they remain hanging in open air. The camera looks slightly downward along the rope from near their shoulder and forearm, but the frame must always retain their hands, rope, and part of their arm or torso. Far below, the white water keeps surging; the falling wood is swallowed immediately by the current, spray breaks into fine droplets, and cold mist lifts upward. They let their eyes drop once toward the void, then immediately drag their head and gaze back toward the broken lip. Their fourth emotional layer is fear seen for a flash and then forcibly contained: eyelids tremble, the eyeballs drop for one instant, lips shake slightly, the inhale catches, then focus re-forms. They say, <d>[English] Don't look down... come on.</d> The first half is lower, like a self-command; the second is shorter and harder. Wind rises from below and lightly shakes damp hair at the temples, the trouser hems, the hem of the top and the straps. Beat 5 — still hanging in midair with no successful support yet. Use a lateral medium close shot at side level, moving with them and the rope as they swing. They turn their body into a pendulum using core, shoulders, knees and hips to drive themselves toward the broken lip. On the first swing they only brush the broken wood with their fingertips, then immediately swing back out into open air again, still fully hanging on the rope with no stable support anywhere on their body. They instantly change grip, tighten their core, bend their knees, roll a shoulder, and load a second swing. Hold the physical effort close here: tightened core, back under strain, stretched strap, forearm tendons standing out, palm friction on rope, fast breathing. Their fifth emotional layer is pain and refusal to give up: after the first failure the brow drops lower, eye corners tighten, nostrils pulse, lips press and release, one mouth corner twitches with pain; before the second effort their gaze locks hard on the broken lip again, sharpening from scattered pain back into focused aggression, and the jaw locks again. They say, <d>[English] One more... come on, move.</d> The rope forms a real loaded arc and rebounds, the palms show small blood streaks and wet abrasion, the broken wood sheds more fragments where they strike it, clothes gather and stretch under swing loads then rebound, and hair whips first then settles against the neck and shoulder. Emphasise throughout that they remain suspended in midair after the first failed attempt. Beat 6 — the first and only moment they truly leave the hanging state. Let the camera push in slightly with their upward drive. On the final swing they reach the highest point, slam a forearm over the broken edge, and only now does a knee successfully hook the remaining wood for the first time. The free hand claws the edge, the core contracts violently, and they drag their full body up, roll half the torso back over the lip, and finally collapse onto the edge, breathing hard. Make it absolutely clear that this is the first genuine contact with the upper edge in the entire video. Their sixth emotional layer is shock after survival and near-collapse: the brows have not fully relaxed, the eyes first pin to the ground and edge, then begin to lose focus half a beat later; nostrils keep flaring fast; the lip line trembles from oxygen debt and pain; the mouth corner twitches faintly; the jaw unlocks little by little. No new dialogue. At the very end, let warm wet backlight from overhead catch the cheek, jaw and neck contour again, while cold bounce from below still lives on the underside of the jaw and inner arms. A few more broken boards drop away behind them to prove the space is still unstable. End without cutting to black, holding on their collapsed position at the broken lip, chest heaving, wet hair hanging down, and the blurred white water still surging behind. overall_soundscape: Continuous environmental bed: humid air, distant insects softened by height, rope strain, bridge wood creaks, and the deep constant roar of the water far below. Emphasise close breath, hand pressure on wet rope, boots on damp planks, faint cloth drag, and their low line <d>[English] Easy... keep moving.</d> Build through plank groans, knot friction, splinters and droplets falling, rope fibres fraying, the bridge rupture, body catch on the side rope, palm scrape against coarse fibre, wind rising from below, whitewater impact far down, pendulum swings, rope tension rise and release, wood edge contact on the failed reach, fragments dropping away, clothes and straps shifting under strain, the final impact of forearm and knee on the edge, fingernails scraping wood, rope rebound, more debris falling, and harsh recovery breathing. No bombastic score, no trailer hits, no extra voices. non_diegetic_music: Keep music minimal and subordinate to the physical environment. If any score is used, make it a restrained low-frequency tension bed that thickens during the hanging section and eases slightly into a shocked, breathless aftermath once they collapse onto the edge. No heroic swell, no horror blast, no melodramatic climax.

5 steps201.4K
MiniMax H315s · 16:9
Inspired by Julianc_AIon X

15-second 16:9 ultra-realistic night alley gunfight. Rain, cyan/magenta neon, wet pavement, steam, crates. @image1 is fighter A (bomber / prosthetic). @image2 is fighter B (yellow mask / rifle). @image3 is the storyboard — photocopy outfits, alley, and beats. Do not describe appearance. Use they / their / them only. Extra attackers are anonymous masked extras. 0–2.5s: Wide whip-pan. Two masked extras emerge from steam. B snaps the rifle up. A drops to a low ready; prosthetic catches neon. 2.5–5s: Hard cut, low muzzle-flash close-up. B fires a burst. Brass arcs through rain. 5–7.5s: A sprints and drives the prosthetic into an extra’s chest. Extra crashes through crates. 7.5–10s: A grapples a second extra — deflect, elbow, throw to wet ground. Weapon skids away. 10–12.5s: B pivots, ejects mag, slams a fresh one, fires as a third extra’s cover is shredded. 12.5–15s: A and B back-to-back in the alley. Last extra drops. Smoke, steam, rain. Hold. No on-screen text, no HUD. AUDIO: rain, gunfire, brass, impacts, servo whir, heavy breath, no lyrics.

6 steps321.1K
Seedance 2.015s · 16:9

Direction, not luck

A clean action clip is directed second by second. The workflows here say where the camera starts, when it moves and what happens on every beat. That is why they hold together from the first frame to the last. Open any one and the whole recipe is there to read.

You are the lead

These are avatar workflows. Upload one clear head-and-shoulders photo and the recipe casts you in the scene. Pick a fight, a chase or a stunt and run it. Want a different scene? Change a line of the prompt and run it again. Every clip here started exactly that way.

Questions

Frequently asked questions

Short AI films that cast you as the lead in a fight, a chase, a heist or a stunt. Each workflow on this page is a recipe other people already ran. You bring one clear photo of your face, and the workflow carries it through every shot of the scene.

No. Every step is already written: the prompt, the model and the settings. Press run and the workflow walks the chain for you. Change the prompt if you want a different scene, or keep it as it is. You only spend credits on the steps that render.

Yes, on the workflows built for it. Duels, sparring matches and buddy scenes ask for two photos, one per character, and keep each face with its own role. A workflow shows how many photos it needs before you run it.