PROMPT LIBRARY
Seedance 2.5 prompts, each with the video it made.
Each prompt is quoted word for word from the model maker's guide or a creator's post, next to the output it produced. Open a card for the full prompt.
ByteDance's Seed team launched Seedance 2.5 on July 31, 2026, on Jimeng AI and Doubao, with API access through BytePlus ModelArk. It generates picture and sound together in one pass, up to 30 seconds per clip, up from 15 in Seedance 2.0, and extends clips over several rounds into videos minutes long. One generation can take up to 30 images, 10 video clips and 10 audio clips as references, and it can edit audio and video at set timestamps. The prompts below are ones creators shared, each shown with its video and linked to its post.
Sources: ByteDance Seed: Introducing Seedance 2.5 · BytePlus: How to write better Seedance 2.5 prompts
Kling 4.0 vs Seedance 2.5 prompts: what we measured on 628 prompts →
Seedance 2.5 prompts that get liked: what 582 posts have in common →
Updated 2026-11-03 · 593 prompts, newest first
Other model pages: GPT Image 2.6 · Nano Banana 2.1 · Kling 4.0 · MiniMax H3 · Wan 3.0
Ostrich with palm fronds runs off a mountain ridge
Anime combat android character sheet with turnaround views
Woman submerged in flooded Seoul subway carriage
Anime magician girl presents gifts over black bars
Man drives white vintage Mercedes through misty forest
Woman walks down a sunlit busy city street
Hero rides armored bike through burning dystopian highway
Superbike rider chases kidnappers SUV on highway
Bearded man with pink heart wand faces trolls
Blue hair martial artist enters dark warehouse spotlight
Blue perfume bottle with droplets at golden sunset
Bodyguard blocks grab and escorts executive to SUV
Man finds carved pumpkin breathing on rainy porch
Two fluffy bunnies share a kiss on sofa
Tearful man aims pistol, then couple embraces in meadow
Woman rewinds cassette and records a tape message
Clown smashes glowing pumpkin with baseball bat
Young woman walks into a county fair at sunset
Woman in halter top fights off suited axe attackers
Cute cream bunny in gingham headband
Crowded train window timeline through history
Woman grills lamb skewers beside pink dragon
Red-haired baker enters glowing bakery at night
Lone figure walks floating palace above clouds
Girl skateboards along a sunny coastal road
Warrior becomes a winged angel in a gothic cathedral
Radiant angel with glowing wings and halo
Swordswoman drinks from jug before inn brawl
Sweaty idol resting on gym mat after session
Woman walks a foggy pumpkin path at midnight
Hamster with pink bow peeks from doorway
Lone fighter in a teal holographic pillar arena
Lollipop student faces off on icy rooftop at night
Silhouetted idol on reflective LED stage
Livestream on an infected night train found footage
Woman in tatami room beside shoji doors
Toddler denies eating sugar to his father
Dim bedroom morning before a JOLTCAN energy drink ad
Witch walks through lantern lit Joseon night bazaar
Kitchen shatters into crystal at her touch
Kitten knight shields dragon egg in snowstorm
Young woman cycles through a sunny Korean alley
Woman chases laundry blown off a clothesline
Woman in elevator sees infected reflection
Woman learns to milk a cow on a farm
Candlelit Korean exorcism ritual with a possessed woman
Nothing matches. Try a shorter word.
Seedance 2.5 prompts worth starting from
The most liked prompt in each category on this page, in full, with what makes it work. Copy one, then change the subject.
Phone camera sketch of friends and a bedsheet ghost
Separate reference images lock each girl's hair and outfit, while the style reference supplies only ink and palette so its pose is not copied. Timed shots with explicit cut and whip pan points keep the handheld phone camera motion readable across eight beats.
References:
@[ref1]: Yuki. Shoulder-length silver-white hair with long bangs, olive-green eyes, small round silver earrings, loose pale camisole, flat deadpan face.
@[ref2]: Sumi. Short black bob with uneven bangs, brown eyes, small gold earrings, cream camisole under an oversized off-white shirt, loud and easily flustered.
@[ref3]: the style reference. Hand-drawn look, line quality and palette only; not its pose or layout.
Visual style: Rough hand-drawn anime sketch, loose black ink lines, flat watercolor washes, muted grey, sage and beige palette, soft indoor daylight. Shot as a phone front camera video: handheld, restless, constantly bobbing and tilting with whoever holds it, quick crash zooms and whip pans.
Avoid: No subtitles. No background music. No slow motion. No other people. Yuki and Sumi never merge or swap faces.
Two friends goof around nonstop with a phone camera through a small apartment, each new spot in front of a different strange painting. Every change of place comes through a whip pan or a cut from the phone itself; nobody stands still. A visible gap always separates their faces. Sound: footsteps, phone rustle, shrieks, laughter, one line. Shot list: 8 shots, about 30 seconds.
Scene 1: Continuous, in front of an orange painting with two huge round owl eyes.
Shot 1 (0-4s): Close two-shot, high angle; the camera bobs quickly with Sumi, who holds the phone with her raised right arm from the top right corner. Yuki springs up into frame from below on screen left and flashes a V sign with both hands, face completely deadpan. Sumi sticks out her tongue and shakes her head side to side, her black bob flying. Cut as Yuki snatches the phone from Sumi's hand; the camera whips left in a blur.
Scene 2: Continuous, in front of a blue-grey blob painting with one black hole for an eye.
Shot 2 (4-8s): Medium shot, eye level; the whip pan lands here, and the camera, now in Yuki's right hand, quickly pans right to follow Sumi. Yuki stays at the bottom left edge of frame, deadpan. Sumi skids across the floor from screen left to screen right in her socks, arms flailing. Cut as Sumi spins and lunges straight at the lens.
Shot 3 (8-11s): Close-up, eye level; the camera, still in Yuki's hand, does a fast crash zoom onto Sumi. Sumi's face is pressed almost against the lens, mouth wide open. Behind her, Yuki does a stiff robot dance with her free arm, still deadpan. Sumi bumps the phone; it tumbles out of Yuki's hand. Cut the instant the phone hits the floor.
Scene 3: Later, in front of a dark green painting of a faded face.
Shot 4 (11-15s): Extreme low angle from the floor; the phone already lies face up, locked-off camera looking at the ceiling. Both girls lean in over the lens from opposite sides, Yuki on screen left, Sumi on screen right. Yuki flips her silver hair forward over her whole face; one big eye peers through a gap. Sumi shrieks and jumps back out of frame, then her hand dives in and grabs the phone. Cut as the phone lifts off the floor.
Scene 4: Continuous, a narrow hallway with a grey wall and a brown smeared painting.
Shot 5 (15-19s): Medium close-up, Sumi holding the phone out in front of her and facing it while she runs backward down the hallway, still running flat out, laughing. Far behind her, at center, Yuki glides after her with her hair still over her face, arms limp, closing the distance. Cut as Sumi swings around a corner; the camera whips right.
Scene 5: Continuous, in front of a beige wall with a round orange painting.
Shot 6 (19-22s): Medium close-up, eye level, handheld; the camera heaves with Sumi's breathing. Sumi, holding the phone in her right hand, presses her back to the wall, panting, eyes darting left and right. Yuki is nowhere in frame. Cut as a white cloth drops in from the top of frame.
Shot 7 (22-25s): Insert, close to the lens; the camera jolts once. A single white bedsheet with two black eye holes lowers from above in front of the lens, held by Yuki's hands.
Yuki's line (flat, deadpan): "Boo."
Shot 8 (25-30s): Wide two-shot, high angle; the camera quickly pulls back, then spins with them. Sumi, phone in her right hand, yanks the sheet with her left; the sheet tangles around both girls and they spin together, laughing, Yuki's hair now neat. Both thrust V signs at the lens. Sumi's index finger is still reaching for the lens when the clip cuts to black as the fingertip covers it. Bald man in kiddie pool beside balcony grill
Framing the clip as an amateur phone recording, with handheld shake and clumsy zooms, plus an explicit ban on film grading, keeps the look grounded. A timed three-beat timeline sequences the collapse with a cause (the cat crossing the railing) before the effect, so the failure reads as a chain of events.
RAW HOME PHONE VIDEO. One continuous 30-second clip recorded on a current smartphone from a facing balcony. Amateur handheld, natural micro-shake, occasional clumsy pinch-zooms in and out. No film look, no cinematic grade, no grain recipe. Overcast gray daylight, peeling block facade, rust, smoke, wet concrete, cheap plastic. Trash metal blasting from the black speaker. Feels like a neighbor filmed it.
Characters
A big bald man with beard and mullet, balcony, grill, Jack Daniel's tap, blue inflatable pool, rubber ducks, speaker, railing and building. Man with receding hair with long sides, gold chain, black SLAYER tank, belly and seated posture.
TIMELINE
0-10s: [Handheld phone wide] The man sits dead-calm in the tiny blue kiddie pool, cigarette in his right hand, living like a king. Serious, unhurried, enjoying. Trash metal pounds. With his free left hand he reaches to the smoking grill, picks up a sausage, smells it, then puts it back exactly where it was. Steam rolls. The phone slowly zooms toward his face, then clumsily zooms back out. He never stands.
10-20s: [Handheld, lazy zoom] His large wife in a floral dress steps out the open door with an empty glass. She fills it from the Jack Daniel's tap with apple juice the same amber color as whiskey, and hands it to him. He takes a long swallow without getting up. She hikes the dress and climbs into the cramped pool with him. They splash. Ducks bob. He stays seated, still smoking, still unbothered. Camera wobbles in on the pour, then out to both of them.
20-30s: [Handheld widening, more shake as it gets worse] A black cat walks out the door and crosses the rusty railing. Only after the cat passes along the railing does the terrace start to crack: hairline fractures spread through the slab. The balcony slowly sags. The inflatable pool tilts; water sloshes over the rim. The couple slide with the tipping pool. The cat jumps out of frame. Concrete splits, railing twists, grill and speaker shift. The terrace gives way and they fall with the water as the clip ends.
AUDIO
Loud trash metal from the speaker the entire time. Grill sizzle, tap pour, splash, groaning metal, cracking concrete, rushing water. Distant courtyard noise. No dialogue. No extra score. No cinematic sound design.
STYLE & QUALITY
Looks like a real phone video from today: auto exposure, slightly crushed highlights, ordinary digital noise, real weight and water physics, wet skin and fabric, collapsing structure with coherent debris.
Hope you like it! 🤘 Seoul subway revenge comedy
Sets the camera language in the first line (handheld, fluorescent, real time), then plays the gag in clear beats and spells out what must stay visible at the end.
Create a live-action cinematic revenge-comedy video set in a Seoul subway station. Use realistic handheld camera movement, cold fluorescent lighting, teal-green accents, and real-time action.
A Korean woman in her 20s with a black bob, sunglasses on her head, a fitted crop top, high-waisted blue jeans, black boots, and a brown shoulder bag descends the subway stairs holding a takeaway coffee. A Korean man with messy dark hair and blonde tips, wearing a dark-green bomber jacket, suddenly hits her from behind and laughs as she stumbles against the handrail. Furious, she chases him onto the platform as he boards a train and mocks her through the closing glass doors.
She punches the glass, creating a dramatic crack and opening. As she confronts him, her coffee cup slips from her hand, falls to the platform, and spills visibly across the floor. She grabs his jacket, forcefully throws him through the air, and sends him crashing into a large metal rubbish bin. The bin rattles and wobbles as he lands inside, looking terrified and humiliated.
She leans toward him and says, "Don't mess around." She then calmly walks away while he remains stuck in the bin.
Maintain identical faces, hairstyles, outfits, and location throughout. Include realistic physics, expressive reactions, cinematic close-ups, subway ambience, and clear white subtitles. Keep the action non-gory, the coffee spill visible, and the final dustbin landing unmistakable. Cafe table talk reshot from thirteen new angles
The prompt locks performance, audio, lighting and tempo of the source video, then varies only camera position and lens through a timed shot list. Measured speech windows, an eyeline rule and an empty sky rule keep lip sync and background clean in every new angle.
Prompt (seedance 2.5 video to video):
ONE FINISHED TAKE, RE-PHOTOGRAPHED
@video1
is one finished take: a single locked-off wide, 19.53 s, with its original audio, a young man at an outdoor cafe table talking to one camera. Do not restage, re-perform, re-time or reinterpret it. Re-photograph the identical event from thirteen new camera positions, as if thirteen extra cameras had been rolling around him in the same moment. Every output frame shows the same body at the same instant of
@video1
at the same speed. Output runs the full 19.53 s. Keep the audio of
@video1
unchanged and in sync.
DO NOT CHANGE
Performance, body, hands and what they hold, head angle, eyeline, lip movement, blink timing, movement speed, action start and end times, lighting, exposure, white balance, colour grade, wardrobe, props, set. The only variable is camera position and lens.
SPEECH WINDOWS (measured from the audio of
@video1
)
SPEAKING: 0.00-1.65, 2.05-4.80, 5.95-6.90, 9.65-11.75, 12.20-12.85, 13.35-13.80, 16.70-16.90, 17.60-18.90.
SILENT, mouth closed and still or chewing: 1.65-2.05, 4.80-5.95, 6.90-9.65 (drinking), 11.75-12.20, 12.85-13.35, 13.80-16.70 (chewing), 16.90-17.60, 18.90-19.55.
Whenever his mouth is visible in a SPEAKING window it moves in lip-sync with the audio of
@video1
. He talks while looking down: from 2.05 to 4.80 his head is down at the laptop and he is speaking. Head down is not silent. Never add speech to a silent window, never mute a speaking one.
EYELINE RULE
He performed to ONE camera, the A-camera: the original locked-off wide, directly in front of him at seated eye level. His eyeline stays on that spot in every shot, apart from the glances down at keyboard, mug and pastry already in
@video1
. He never looks for or turns toward a new camera. From above he reads cap and brow; from the sides, three-quarter or profile; from behind, the back of his head. Single exception: the on-axis shot at 00:01-00:02 sits where the A-camera was, so there alone he looks into the lens.
TEMPO RULE
Real time, 1:1 with
@video1
. No speed ramps, added slow motion, freeze frames or time remaps. Exactly one slowed passage exists, already baked into the source: the pastry at the top of its arc, 12.5-13.55. Reproduce it at the speed it already has, no slower, no longer.
TOSS TIMING
The pastry leaves his fingers at 12.43 and is caught at 13.60: airborne about one second only. Do not stretch, loop or hang the flight. Before 12.43 and after 13.60 it is in his hand.
EMPTY SKY
On every low or upward-tilting shot the sky is plain pale overcast and empty. No skyscrapers, glass towers, solar panels, cranes, birds, aircraft, sun disc, structured clouds, lens flare or signage. Above his head only cream wall, white pillars and flat pale sky.
CAMERA PLAN
[00:00-00:01] High-angle MCU from above and to his right, about 40 degrees down, chest up, slow push-in, 35mm feel, moderate depth of field, cap brim and open gesturing palms, marble edge in the lower foreground. Speaking, reading three-quarter, not to this lens.
[00:01-00:02] ON THE A-CAMERA AXIS. ECU of his face, dead front, seated eye level, 85mm feel, shallow focus, pillars soft cream behind. The one shot where he looks into the lens. Hands drop below frame onto the keyboard; mouth closes at 1.65.
[00:02-00:05] Macro along the MacBook keyboard to his typing fingers, lens centimetres above the keys, razor-thin focus, keycaps sharp, his face soft far behind, head down. Speaking 2.05-4.80: where his mouth shows in the blur it moves with the audio. Fingers lift at 5.0.
[00:05-00:06] Top-down bird's eye straight down over the table, 24mm feel, deep focus, locked off. Marble disc fills frame: open laptop, white mug, financier, his cap at the bottom edge. Right hand crosses to the mug handle.
[00:06-00:07] Lens resting on the marble at table height, 50mm feel. White mug large and sharp in the left foreground; his hand enters from the right, grips the handle, lifts the mug up out of frame. Torso soft behind. Speaking 5.95-6.90.
[00:07-00:09] First-person POV from his seat, ultra-wide 18mm feel, gentle handheld breathing. The mug rim rises into the bottom of frame and tips toward the lens; laptop screen and marble beyond in deep focus. No face; framing dips to the mug, never up. Silent.
[00:09-00:10] Extreme wide, slow dolly out, 24mm feel, deep focus. He is small at his table mid-terrace: cream wall, pillars, concentric-circle mural, taped boxes behind, slatted bench visible. He sets the mug on the marble 9.4-10.0; speech resumes 9.65.
[00:10-00:12] Low and to his left, lens just above the marble, 50mm feel, shallow focus, table edge and mug soft in the near foreground. Right hand reaches out at 10.5 and lifts the golden financier into sharp focus before his left profile. Talking to the A-camera: full profile here.
[00:12-00:14] Low beside him, 35mm feel, slight Dutch. Opens on the pastry in his fingers; at 12.43 it leaves his hand and the camera tilts UP, following it against plain pale overcast sky and the top of a white pillar through the baked-in slow apex 12.5-13.55, then tilts DOWN for the catch at 13.60 into cupped hands, head tipped back, mouth open, then head coming forward.
[00:14-00:15] Behind him, over his right shoulder, 50mm feel, moderate focus. Back of the black Yankees cap and long dark hair fill frame-left; laptop glow and marble past his shoulder. Head settles level, hands bring the pastry toward his mouth. No face. Silent.
[00:15-00:16] ECU on his mouth and jaw, 100mm macro feel, very shallow focus, soft open-sky light. The financier enters from below, he bites, crumbs, the jaw chews. Chewing, not speaking. Nothing above the nose in frame.
[00:16-00:18] Slow orbit from his left toward front, 35mm feel, moderate focus, mural and pillars parallaxing behind, table crossing the foreground. He chews, speaks 16.70-16.90 and from 17.60, open palms, a small shrug. As the orbit nears front his gaze lands near, not into, this moving lens.
[00:18-00:20] Top-down bird's eye straight down over the table, 24mm feel, deep focus, locked off. Hands come down beside the laptop and settle, cap centred, speech ends 18.90, then still. Hold to 19.53.
SET AND SUBJECT CONTINUITY (identical in every shot)
Young man, long dark hair, black short-sleeve button-up shirt with a small chest logo, black New York Yankees baseball cap, grey trousers. Light wood slatted bench, round white marble table, outdoor cafe terrace. On the table: open dark grey MacBook, white mug with a circular graphic, one small golden financier. Behind him: cream wall, white cylindrical pillars, large black-and-white concentric-circle mural, taped cardboard boxes. Soft overcast open-shade daylight, no hard shadows, constant direction and softness. Warm bright grade: cream walls, golden wood, neutral white marble, warm skin, deep neutral blacks. He is alone; no passers-by, no staff.
No on-screen text, captions, subtitles, watermarks or logos beyond the small chest logo and cap emblem already in
@video1
. 1950s diner: time freezes, then rewinds
A single effect carried for 30 seconds. The diner is described once in detail so the freeze and rewind read clearly against it.
Photorealistic cinematic 1950s American diner, chrome stools, red vinyl, neon glow and checkerboard floor, shot with modern lived-in realism and soft natural window light. Subtle handheld texture, warm practicals, rich period detail, heavy film grain.
0-4s: [Medium Wide] A striking young woman in her early 20s sits alone at the counter, calm and slightly amused, slowly sipping a tall thick milkshake through a straw. Behind her a young waitress in classic uniform approaches with a tray of eggs and bacon in one hand and a full glass coffee pot in the other. An older lady starts rising from a nearby booth.
4-8s: [Dynamic Tracking] The older lady collides hard into the waitress. Tray, plate, eggs, bacon and coffee pot explode upward in chaotic slow motion. Coffee erupts into long liquid ribbons and perfect suspended droplets. Camera immediately begins a smooth continuous orbit around the impact. Time locks completely at the peak of the spill. Every face freezes in pure shock. Only the girl at the counter keeps moving, completely unfazed.
8-17s: [Slow 360° Orbital] Camera glides in a full elegant orbit through the frozen diner. Coffee hangs in mid-air as glassy ribbons and spheres with perfect volume and surface tension. Bacon strips, eggs and the spinning tray float weightlessly. Patrons and waitress remain locked in startled expressions. The girl at the counter takes one slow, deliberate sip, eyes half-lidded, almost bored, while the entire frozen world (except her) begins an elegant reverse: every droplet, every piece of food and every person rewinds smoothly back to the exact starting positions.
17-24s: [Medium Shot] Rewind lands perfectly. Waitress stands balanced again with tray and coffee pot. The girl lifts her eyes, raises two fingers in a small casual gesture and softly calls the waitress by name. The waitress turns toward her just before the older lady begins to stand, completely avoiding the collision. A tiny private smile crosses the girl’s face.
24-30s: [Extreme Close-Up] Hard cut to her face as she takes one last slow sip. Soft knowing smile, eyes almost closed in quiet satisfaction, like she has done this a hundred times. Shallow depth of field, creamy bokeh of the neon diner behind her.
Photorealistic, ultra-detailed fluid physics, perfect motion blur only on moving elements, stable characters, cinematic lighting, heavy natural film grain, no artifacts, movie-level temporal coherence, high rewatch value. Lab containment breach, sci-fi horror
Asks for an immediate hook and grounded practical effects, which keeps the creature from looking like CGI.
Cinematic sci-fi horror-action short film set in a research laboratory, opens with an immediate strong hook, natural realistic pacing, grounded practical-effects style. Lab filled with steel equipment, glass containment tubes, monitors, specimen tables, harsh fluorescent lighting. Five scientists in lab coats work at various stations. Natural handheld camera work, muted authentic color grading, all sound fully natural and diegetic — no music, no artificial sound design until the tension score kicks in.
>
> **[0-1s]** Hard cut: a containment tube shatters violently, glass exploding across the lab floor.
>
> **[1-2s]** A small dark parasite creature skitters out fast, latching instantly onto a scientist's neck.
>
> **[2-3s]** He screams, collapsing to his knees, the others recoiling in shock.
>
> **[3-4s]** Dark veins pulse rapidly beneath his skin, visible even through his collar, his body convulsing.
>
> **[4-5s]** His eyes turn a deep obsidian black — he rises, movements jerky and unnatural.
>
> **[5-6s]** He lunges at the nearest colleague, parasite tendrils extending from his mouth toward her.
>
> **[6-7s]** She's caught, tendrils latching onto her neck, her scream cutting short.
>
> **[7-8s]** Veins spread fast across her too, body convulsing against a lab table, equipment crashing down.
>
> **[8-9s]** She rises seconds later, eyes black, joining him, both turning toward the remaining three.
>
> **[9-10s]** The three scientists scramble back, one grabbing a fire axe from an emergency case on the wall.
>
> **[10-11s]** The two infected charge fast, jerky inhuman movements, tendrils whipping from their jaws.
>
> **[11-12s]** The lead scientist swings the axe hard into the first infected — a solid impact, it staggers into a specimen table, equipment shattering.
>
> **[12-13s]** A second scientist grabs a fire extinguisher, blasting the other infected in the face, it recoils, tendrils flailing.
>
> **[13-14s]** The third scientist shoves a steel cart into the staggering infected's path, sending it crashing down.
>
> **[14-15s]** Both infected collapse motionless, twitching, tendrils retracting slowly.
>
> **[15-16s]** The three survivors regroup, breathing hard, scanning the wrecked lab for more movement.
>
> **[16-17s]** Wide shot: broken glass and toppled equipment everywhere, alarm lights beginning to flash red.
>
> **[17-18s]** A third containment tube behind them starts cracking, unnoticed by the group.
>
> **[18-19s]** It shatters suddenly — another parasite creature launches out fast.
>
> **[19-20s]** The lead scientist spins just in time, swinging the axe and slicing it out of the air mid-leap.
>
> **[20-21s]** It hits the floor, motionless, dark fluid spreading slightly beneath it.
>
> **[21-23s]** The group presses their backs together, weapons raised, scanning every corner of the lab.
>
> **[23-25s]** Emergency lights flash steadily red, alarm klaxon blaring through the facility.
>
> **[25-27s]** The lead scientist gestures toward the exit; they move together cautiously, stepping over debris and fallen colleagues.
>
> **[27-29s]** They reach the heavy lab door, the lead scientist sealing it shut behind them with a hard mechanical clang.
>
> **[29-30s]** Final shot: the three survivors lean against the sealed door, breathing hard, alarm lights flashing red across their exhausted faces. Cut to black.
>
> Natural diegetic sound throughout — glass shattering, guttural convulsions, screaming, metal impacts, alarm klaxon, heavy breathing — layered with a tense instrumental score building through the fight and easing at the end. No text overlay, no watermark. Natural handheld camera with realistic shake during action, muted clinical color grading. How to write prompts for this model
Open with one sentence that sums up the video
ByteDance's BytePlus guide asks for the main idea first, in the shape subject, location, event, style and special camera movement, before any scene detail. Its template then runs reference mapping, main idea, timeline, consistency and constraints, with the essential constraints last, and tells you to include only the parts your project needs. Source: BytePlus: How to write better Seedance 2.5 prompts
Use labeled blocks and say how the camera is held
In our study of 582 creator prompts matched with their likes, 61% of the most liked quarter break into labeled blocks such as CHARACTERS:, against 48% of the least liked, and 49% ask for a handheld camera, against 34%. Time ranges and a 30-second plan appear just as often in both, so treat them as the floor rather than the edge. Source: AgentCraft: Seedance 2.5 prompts that get liked
Split the clip into time ranges
ByteDance's own example prompts set the pacing with ranges like 0 to 5 seconds and give each range its action. With 30 seconds in one take, creators do the same: 311 of the 593 prompts here give time ranges, and 112 of the 162 that start from references. Source: ByteDance Seed: Introducing Seedance 2.5 · AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data
Tag every reference and say what it is for
In ByteDance's examples each upload is named at the start with its role, such as @Image 1 as the venue or a clay render as the camera path. 66 of the 162 reference prompts on this page number their uploads this way, so the model knows which face, place or motion to take from where. Source: ByteDance Seed: Introducing Seedance 2.5 · AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data
Write the camera as a shot list
Name the move: tracking, orbit, handheld, or a single continuous take, as ByteDance's examples do. 494 of the 593 prompts here use camera terms, and 302 split the prompt into labeled blocks such as CAMERA, LOOK and SCENES. Source: ByteDance Seed: Introducing Seedance 2.5 · AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data
Write the sound and the spoken lines
Seedance 2.5 makes audio and picture in the same pass and keeps them in sync, so the soundtrack is part of the prompt. 342 prompts here describe sound, music or a voice, and 220 quote a line to be spoken or shown. Source: ByteDance Seed: Introducing Seedance 2.5 · AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data
State the length and the frame
390 prompts here state a length in seconds and 270 an aspect ratio, often on one line such as 30 seconds, 16:9. A single pass goes up to 30 seconds; for longer pieces, extend the clip in rounds. Source: AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data · ByteDance Seed: Introducing Seedance 2.5
When extending or editing, say what stays
ByteDance's extension examples list what to keep consistent: characters, scene, style and sound. Its edit examples name the one thing to change, such as adjusting only the camera movement. 307 prompts here say keep, preserve or maintain. Source: ByteDance Seed: Introducing Seedance 2.5 · AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data
Keep complex motion simple
The Seed team says complex motion and scenes where several subjects interact still need work. When a fight or a crowd scene breaks, give each time range one main action and fewer people at once. Source: ByteDance Seed: Introducing Seedance 2.5
Questions about this model
Is there an official Seedance 2.5 prompt template?
Yes. ByteDance's BytePlus guide gives five parts in order: reference mapping, main idea, timeline, consistency and constraints. Time instructions work to the whole second, such as 0 to 3 seconds, and for edits it uses verbs like add, remove, replace and change to. Source: BytePlus: How to write better Seedance 2.5 prompts
What do the most liked Seedance 2.5 prompts have in common?
Labeled blocks and a handheld camera. Across 582 prompts from X, the most liked quarter uses labeled blocks 61% of the time against 48%, and handheld 49% against 34%; they are also longer, a median of 2,788 characters against 2,069. Words like cinematic appear just as often in the least liked. Source: AgentCraft: Seedance 2.5 prompts that get liked
What is Seedance 2.5?
ByteDance's video generation model, launched by its Seed team on July 31, 2026. It generates video with matching audio from text and from image, video and audio references. Source: ByteDance Seed: Introducing Seedance 2.5
How long can a Seedance 2.5 video be?
Up to 30 seconds in a single pass, twice the 15 seconds of Seedance 2.0. Extending over several rounds produces videos several minutes long. Source: ByteDance Seed: Introducing Seedance 2.5
How many references can one generation use?
Up to 30 images, 10 video clips and 10 audio clips in a single pass. Source: ByteDance Seed: Introducing Seedance 2.5
Does Seedance 2.5 generate sound?
Yes. It generates audio and video jointly, and ByteDance says the two stay in sync. Write dialogue, music and sound effects in the prompt. Source: ByteDance Seed: Introducing Seedance 2.5
Where can I use Seedance 2.5?
In Jimeng AI and Doubao, under video generation, where you select Seedance 2.5. Developers get API access through BytePlus ModelArk. Source: ByteDance Seed: Introducing Seedance 2.5
How long are Seedance 2.5 prompts?
Long. The 593 prompts on this page have a median of 2,378 characters, about 344 words over 12 lines. Prompts built on references run longer, with a median near 3,000 characters, because they also assign each upload its role. Source: AgentCraft: Nano Banana 2.1 and Seedance 2.5 prompt data