All Tools

Seedance 2.5 Prompt Generator

Paste your rough video idea. Get a production-ready Seedance 2.5 prompt with timestamps, staged structure, audio syntax, and camera direction — optimized for Dreamina.

1. Your Video Idea

0/2000

2. Parameters

10s · 2–3 shots
5s20s30s

Free · No sign-up needed · Powered by Claude

Before & After Examples

Before

a potter making a cup

After

A ceramic artist finishes a pale blue cup in a studio at dawn, lifts it from the wheel, and places it in the center of a wooden shelf. Soft morning light enters through the window. The wet clay has a delicate sheen, and the workbench remains tidy. Begin with a medium shot of the wheel-throwing process, slowly push in toward the cup's surface texture, then cut to a frontal view of the shelf. Retain the low hum of the pottery wheel, the friction of clay, and subtle indoor ambience.

Added: specific action sequence, lighting direction, material texture, camera moves, audio details. Style described visually, not with mood words.

Before

café opening in the morning

After

A barista opens the shutter of a small corner café at dawn, wipes down the counter, and starts the first pour of the morning. Warm low sunlight through the front glass, steam rising, visible sensor grain. Medium shot, slow push-in toward the cup, then cut to the street. (Soft, rhythmic piano music plays in the background) <The metal shutter rattles upward> <Steam hisses from the wand> Dialogue language: American English. The barista says in natural, conversational American English: {First one's always the best one.}

Added: specific actions, visual style as observable details, camera treatment, full audio syntax — ( ) music, < > SFX, { } dialogue with language tag.

Before

show a sneaker product video

After

0-5 seconds: A matte white sneaker with neon green accents sits on a glossy black turntable, rotating slowly under hard studio spotlight. Camera orbits at eye level. Clean white background. End state: sneaker has completed one full rotation. 5-10 seconds: Camera pushes in to extreme close-up on the sole tread pattern, then rack focus to the knit texture of the upper. End state: upper texture fills the frame, sharp and detailed. (Subtle bass-heavy electronic pulse) <Soft turntable motor hum> Hyperrealistic product photography style, 8K detail.

Added: timestamps with end states, product details, turntable motion, camera orbit + push-in + rack focus, studio lighting, audio syntax, style reference.

The Core Formula for Seedance 2.5

Seedance is unusually literal — it does what you ask, but only if you say it. Most bad output is an underspecified prompt, not a model failure.

1

Subject & Action

Who or what, and what they do. Seedance is unusually literal — it does what you ask, only if you say it. Be specific about appearance, clothing, material.

ceramic artist in dark green apronmatte black folding desk lampgolden retriever puppybarista in white apron
2

Scene & Environment

Location, time of day, weather, spatial relationships, background state. Specific details beat mood words.

pottery studio at dawnrain-soaked Tokyo side street at nightsmall corner café with front glass windowtechnology gallery with curved walls and gray floor
3

Visual Style

"Cinematic" leaves the model free. Describe what you actually see: lighting direction, color, materials, grain, texture.

soft morning light through window, visible sensor grainwarm low-angle sunlight, long shadowsbright realistic museum-documentaryraw handheld footage, no colour grading
4

Camera & Framing

Shot size, angle, movement, and focus subject. For niche terms, also describe the visible result.

medium shot, slow push-inextreme close-up on surface texturelocked-off wide shotsteadicam tracking from the front
5

Audio & Dialogue

Music in ( ), sound effects in < >, dialogue in { }. Name the language before non-Chinese dialogue.

(soft rhythmic piano music)<metal shutter rattles upward><steam hisses from the wand>{First one's always the best one.}
6

End State

What is directly visible when the shot ends. Positions, prop ownership, scene state — never a mood.

cup centered on wooden shelfbouquet lies flat on workbench with ribbon bow facing camerahand has left frame, only white plate remainsboth characters stand behind workbench

Before & After Prompt Examples

See how vague prompts become production-ready Seedance 2.5 prompts with end states, audio brackets, and staged structure.

Before

a potter making a cup

After
A ceramic artist finishes a pale blue cup in a studio at dawn, lifts it from the wheel, and places it in the center of a wooden shelf.

Soft morning light enters through the window. The wet clay has a delicate sheen, and the workbench remains tidy.

Begin with a medium shot of the wheel-throwing process, slowly push in toward the cup's surface texture, then cut to a frontal view of the shelf.

Retain the low hum of the pottery wheel, the friction of clay, and subtle indoor ambience.

Added: specific action sequence, lighting direction, material texture, camera moves, audio details. Style described visually, not with mood words.

Before

café opening in the morning

After
A barista opens the shutter of a small corner café at dawn, wipes down the counter, and starts the first pour of the morning.

Warm low sunlight through the front glass, steam rising, visible sensor grain. Medium shot, slow push-in toward the cup, then cut to the street.

(Soft, rhythmic piano music plays in the background)
<The metal shutter rattles upward>
<Steam hisses from the wand>
Dialogue language: American English. The barista says in natural, conversational American English: {First one's always the best one.}

Added: specific actions, visual style as observable details, camera treatment, full audio syntax — ( ) music, < > SFX, { } dialogue with language tag.

Before

show a sneaker product video

After
0-5 seconds: A matte white sneaker with neon green accents sits on a glossy black turntable, rotating slowly under hard studio spotlight. Camera orbits at eye level. Clean white background. End state: sneaker has completed one full rotation.

5-10 seconds: Camera pushes in to extreme close-up on the sole tread pattern, then rack focus to the knit texture of the upper. End state: upper texture fills the frame, sharp and detailed.

(Subtle bass-heavy electronic pulse)
<Soft turntable motor hum>
Hyperrealistic product photography style, 8K detail.

Added: timestamps with end states, product details, turntable motion, camera orbit + push-in + rack focus, studio lighting, audio syntax, style reference.

Before

a 30 second flower shop video

After
[Generation Goal]
Generate an instructional video showing a flower shop's order-packing process. <Florist> and <Store Assistant> arrange, wrap, and hand off a bouquet together.

[Stage 1]
Initial state: <Florist> stands behind the workbench. Loose flower stems, scissors, and wrapping paper lie on the tabletop.
Primary event: <Florist> arranges the stems and trims them to length.
End state: <Florist> holds the bouquet in the left hand, and the scissors are back on the right side of the workbench.

[Stage 2]
Continue from the previous stage: both characters retain the same identities and clothing, and <Florist> still holds the bouquet.
Primary event: <Store Assistant> unfolds the wrapping paper. <Florist> places the bouquet inside and ties it with a green ribbon.
End state: the wrapped bouquet lies flat in the center of the workbench, with the ribbon bow facing the camera.

[Stage 3]
Primary event: <Store Assistant> picks up the bouquet and places it on the pickup shelf.
End state: the bouquet is centered on the pickup shelf, and both characters stand behind the workbench inspecting the finished order.

[Maintain Consistency]
Keep <Florist> and <Store Assistant>'s identities, clothing, workbench orientation, scissors position, and bouquet ownership consistent.

Used staged video structure for 30s content. Each stage has one primary event + observable end state. Consistency block at the end prevents character/prop drift.

Before

make a dragon scene

After
A massive obsidian dragon with iridescent scales perches on a cliff edge above a cloud-filled valley at dusk. It spreads its wings slowly — wingspan fills the frame.

0-5 seconds: Camera crane up from below reveals the full scale of the dragon. Volumetric fog swirls around its talons. End state: full dragon visible against the orange sky.

5-10 seconds: The dragon inhales, chest glowing orange, then exhales a torrent of blue fire into the sky. Camera tracks the flame upward. End state: fire arc reaches its peak, dragon silhouetted against the glow.

<Deep rumbling roar echoes through the valley>
<Fire crackles and hisses>
(Epic orchestral swell builds through the sequence)

Added: dragon appearance details, timestamp ranges with end states, specific camera moves, sound effects in < > brackets, music in ( ) brackets.

Audio & Dialogue Syntax

Bracket Types

( )

Music

(Soft, rhythmic piano music plays)

< >

Sound Effects

<A bell rings in the distance>

{ }

Dialogue

The barista says: {First one's always the best.}

Dialogue Formula

Dialogue language: American English.
The barista says in natural,
conversational American English:
{First one's always the best one.}

Name the language before the line. Include delivery style (softly, natural, flatly) and the speaker identity.

Suppress Audio

Use (no music) or (no BGM) to suppress background music. Use (no subtitles) to prevent burned-in text.

Camera & Style Cheat Sheet

Camera Moves

static shotlocked-off shotpush inslow push-inpull outpan leftpan rightwhip pantilt uptilt downlateral movetracking shotfollow shotorbit shotorbit 360°crane upcrane downdolly zoom

Lighting

golden hourwarm low-angle sunlightsoft morning lightdawn light through windowhard studio spotlighttop-down spotlightsoft diffused lightovercast flat lightbacklight silhouetterim lightneon glowpractical lightscandlelightmoonlightcold blue lightvolumetric light raysgod raysstudio three-point lighting

Visual Styles

photorealistichyperrealisticcinematic35mm filmanamorphicIMAXdocumentaryraw handheld footagevisible sensor grainno colour gradinganimeStudio GhibliPixar 3Dwatercoloroil paintingcharcoal sketchclaymationstop motion

Shot Budget & Structure

5s
1 shot

Single continuous shot. Describe one primary action with a clear end state.

10s
2–3 shots

Use timestamp ranges (0-3s, 3-7s, 7-10s). Each range needs an end state.

15–20s
3–5 shots or 2–3 stages

Consider [Stage] structure. One primary event per stage.

25–30s
3–5 stages

Use [Stage 1], [Stage 2], etc. Each stage: one event + observable end state + consistency carry-forward.

Timestamp Format

0-5 seconds: [Shot 1]. End state: ...
5-10 seconds: [Shot 2]. End state: ...
  • Integer-second ranges, consecutive, non-overlapping.
  • Each range is a time budget, not a frame-accurate edit point.
  • Every shot needs a directly visible end state.

Staged Video (20-30s)

[Generation Goal]
Generate a <type> video...

[Stage 1]
Initial state: ...
Primary event: ...
End state: ...

[Maintain Consistency]
Keep character identity...
  • One primary event per stage.
  • End states must be directly visible — not moods.
  • Consistency block at the end prevents drift.

Common Mistakes to Avoid

Don't put generation parameters in the prompt — aspect ratio and duration are set separately

Don't use mood words as style ("cinematic", "vibrant") — describe what you see: lighting direction, color, grain, texture

Don't write "@Images 1 through 4 define four characters" — bind each reference individually

Don't skip exclusions — if a reference contains a background or person you don't want, say "Do not use the image background"

Don't put multiple primary events in one stage — one state change per stage prevents dropped beats

Don't describe end states as moods — "feels peaceful" fails, "cup centered on shelf, hand has left frame" works

Don't overlap timestamp ranges — 0-5s then 4-9s breaks, use 0-5s then 5-10s

Don't demand frequencies — "complete three actions in one second" never works

Don't re-describe motion a reference video already carries — it conflicts with the reference

Reference Binding Syntax

When using images, videos, or audio as references, bind each one individually and state what to use — and what NOT to use.

Wrong

@Images 1 through 4 define four
characters respectively.

Does not say which image is which character.

Right

@Image 1 defines the artist's facial
features and dark green apron.
Do not use the image background.

@Image 2 defines the wooden workbench
and morning light of the studio.
Do not use the people in the image.

Each reference bound individually with exclusions.

Frequently Asked Questions

What is Seedance 2.5?+
Seedance 2.5 is a video generation AI model developed by ByteDance, available through the Dreamina platform. It generates videos up to 30 seconds from text prompts, images, or multi-reference materials. It supports staged video structure, reference binding with @Image/@Video/@Audio syntax, and native audio generation.
How do I write timestamps in Seedance 2.5 prompts?+
Use integer-second ranges like "0-5 seconds:", "5-10 seconds:". Ranges are a time budget — actions may occur slightly before or after a boundary. Keep ranges consecutive and non-overlapping. For videos over 20 seconds, use the [Stage] structure instead of timestamps.
What is the staged video structure in Seedance 2.5?+
For videos 20-30 seconds, divide content into stages using [Generation Goal], [Stage 1], [Stage 2], etc., and [Maintain Consistency] blocks. Each stage gets one primary event and a directly visible end state. The consistency block at the end lists character identity, clothing, prop ownership, and spatial relationships to preserve across stages.
How do I add audio to Seedance 2.5 prompts?+
Use bracket syntax: ( ) for music — e.g. (Soft piano music plays). < > for sound effects — e.g. <A bell rings>. { } for dialogue — e.g. {Hello, welcome back.}. For non-Chinese dialogue, name the language before the line: "Dialogue language: American English. The speaker says: {text}".
What durations does Seedance 2.5 support?+
Seedance 2.5 supports videos from 5 to 30 seconds. For 5s, use a single continuous shot. For 10s, use 2-3 timestamp ranges. For 15-20s, use timestamps or stages. For 25-30s, use the [Stage] structure with 3-5 stages for best results.
Is this prompt generator free?+
Yes, the prompt optimizer is completely free. You need to sign in with an inReels account (also free), but no credits are charged. You can optimize as many prompts as you want and use them on Dreamina or any platform that supports Seedance 2.5.
How is Seedance 2.5 different from Minimax H3?+
Key differences: Seedance 2.5 supports up to 30 seconds (H3 max 15s). Seedance uses integer-second timestamps (0-5s) while H3 uses MM:SS.mmm format. Seedance has bracket-based audio syntax — () music, <> SFX, {} dialogue. Seedance supports up to 50 reference materials with @Image/@Video/@Audio binding. Seedance has staged video structure for longer content.

Write Prompts That Get Results

Stop guessing. Paste your idea, get a production-ready Seedance 2.5 prompt in seconds.

Try inReels Free — No Credit Card

Free prompt optimizer + 15 free credits for video generation

Read the full Seedance 2.5 prompt engineering guide →

Questions? Chat with us

We typically reply instantly