ByteDance AI video model with native audio
Seedance 2.5 AI Video Generator
Seedance 2.5 turns one written shot note into a finished cinematic clip with its own sound, from a marble lion bursting into golden sand in 1080p to a 15 second single take with no cuts.
Seedance 2.5 is ByteDance's video model for text, first frame and reference inputs. It makes single takes from 4 to 30 seconds, speaks dialogue in 11 languages, and can follow up to 50 reference files (30 images, 10 videos, 10 audio). We made every clip on this page on Pixel Dojo from a text prompt alone, and each one comes with its exact prompt so you can start from it.
30 credits ($1.88) per 5 s clip at 720p with audio. Included in every plan, from $10/month.
Already a member? Sign in
By Brian Lovett · Updated
What you can make with Seedance 2.5, prompts included
Seven clips in seven genres, each made on Pixel Dojo from a single text prompt with sound generated alongside the picture: a VFX reveal, a 15 second single take, spoken dialogue, action, fashion, aerial and 3D animation. We kept every prompt exactly as we wrote it. Tap Use this prompt to start from any of them.
VFX transformation in 1080p. A marble lion cracks along glowing gold seams, the shell bursts apart and a glowing golden lion emerges from a swirling cloud of sand, its face steady while stone and dust fly past. Start from this prompt for product reveals, title openers or any before and after moment.
Medium wide shot with a slow push in. A white marble lion statue on a plinth in a dark, empty gallery cracks along glowing gold seams, the stone shell bursts apart in slow motion and a living lion made of swirling golden sand shakes out its mane and roars, dust hanging in a single shaft of skylight, 35mm lens, rich contrast. Sound: stone cracking, a rushing sand whoosh and a deep echoing roar.
One continuous 15 second shot for 45 credits. We wrote it as three 5 second beats and all three play out in order with no cuts: she walks up the snowy platform, the train pulls in behind her as the camera comes around to her face, then she steps aboard and looks back through the window of the closing door. Her face, red coat, scarf and suitcase stay the same from the first frame to the last.
One continuous 15 second Steadicam shot with no cuts, 35mm lens, soft film grain. A woman in her thirties with a dark bob, a long red wool coat and a cream scarf pulls a small tan leather suitcase along an empty snowy train platform at blue hour, snow falling through the warm glow of old iron lamps. 0-5 seconds: the camera tracks beside her as she walks toward us and stops at the platform edge, the suitcase upright at her right side. 5-10 seconds: a dark green passenger train glides in behind her, its headlight flaring and a swirl of snow lifting off the platform, and the camera arcs around to her face as she looks up at it. 10-15 seconds: the doors slide open onto a warmly lit carriage, she lifts the suitcase and steps aboard, then turns back toward us once as the doors close. Keep her face, red coat, scarf and the single suitcase the same throughout. Sound: wind, boots crunching on snow, suitcase wheels, the long hiss of train brakes and the thud of the closing doors.
Dialogue with native audio in 1080p. She sets down a steaming bowl, looks up with a warm smile and says the English line word for word: "Last bowl of the night. Eat it while it's hot." Steam, lantern glow and rain on the glass finish the scene. Put your line in curly braces and name the language first, as this prompt does.
Medium close-up, slow push in. A noodle bar chef in her fifties with silver-streaked hair in a low knot and a black linen apron lifts a steaming bowl onto the counter, looks up at the customer with a small smile and says warmly in American English: {Last bowl of the night. Eat it while it's hot.} Warm paper-lantern light, steam curling through the frame, a rain-streaked window behind her, 50mm lens, shallow depth of field, fine film grain. Sound: bubbling broth, a sizzling pan, soft rain on the glass.
Action with real physics. A black horse gallops through shallow surf against a low golden sun, spray kicking up with every stride and its mane and tail streaming back, while the camera races alongside in one continuous shot. A bold silhouette like this reads instantly at thumbnail size.
Low tracking shot racing alongside a black horse at full gallop through shallow surf at sunset, no rider, spray exploding from every hoof and its long mane and tail streaming back, the low golden sun behind it turning each droplet into a spark of light, one continuous shot with no cuts, 50mm lens, slight motion blur, warm film grain. Sound: pounding hooves in water, crashing waves and heavy breathing.
Fashion and dance in a full 9:16 frame for vertical feeds. A dancer with a platinum buzz cut spins under one hard spotlight as wide arcs of crimson silk sweep out and throw sharp shadows across the white floor. Swap in your own wardrobe and color to make it yours.
Full-body shot from a low angle with the camera slowly circling. A dancer with a platinum buzz cut in a flowing crimson silk suit spins at the center of a seamless white studio, the long jacket flaring out in wide arcs, a single hard overhead spotlight casting a sharp circular shadow on the floor, 24mm lens, glossy fashion editorial grade with deep saturated reds. Sound: a driving house beat and the snap of silk on every turn.
Aerial landscape in widescreen 21:9. The drone glides low over a pale braided glacial river on black sand, then tilts up to jagged snowy peaks glowing pink at sunrise. Use it as an opener, a travel reel or a title background.
Aerial drone shot gliding fast and low over a glacier-blue river braiding through black volcanic sand at sunrise, then tilting up to reveal a jagged snow-capped mountain range glowing pink, wisps of mist curling off the water, 16mm wide lens, crisp high dynamic range color. Sound: rushing wind and the low roar of the braided river.
Stylized 3D animation with a feature film look. A round teal robot on a moonlit ledge reaches up to catch a drifting paper lantern, and its amber eye brightens as it does. Cool blue moonlight against warm lantern glow gives it a storybook feel.
Stylized 3D animated feature look, medium shot tracking sideways. A small round robot with a glossy bright teal shell, stubby legs and one big glowing amber eye tiptoes along a moonlit rooftop garden ledge, reaches up with a thin arm to catch a drifting paper lantern, and its eye brightens. Soft blue volumetric moonlight, warm orange lantern glow, painterly textures, gentle depth of field. Sound: tiny servo whirs, soft footsteps on stone and crickets.
What Seedance 2.5 does best, and how to get clips like these
Draft mode is built into Seedance 2.5: preview any shot at 480p for 15 credits, then upgrade that exact shot to 1080p with the same framing, motion and spoken line. The lion and chef above started as these two previews. Below them are the techniques behind every clip on this page.
480p preview for 15 credits. Upgraded, it became the 1080p lion above, with the same framing and timing and sharper detail.
Medium wide shot with a slow push in. A white marble lion statue on a plinth in a dark, empty gallery cracks along glowing gold seams, the stone shell bursts apart in slow motion and a living lion made of swirling golden sand shakes out its mane and roars, dust hanging in a single shaft of skylight, 35mm lens, rich contrast. Sound: stone cracking, a rushing sand whoosh and a deep echoing roar.
480p preview for 15 credits. Upgraded, it became the 1080p chef above shot for shot, speaking the same English line.
Medium close-up, slow push in. A noodle bar chef in her fifties with silver-streaked hair in a low knot and a black linen apron lifts a steaming bowl onto the counter, looks up at the customer with a small smile and says warmly in American English: {Last bowl of the night. Eat it while it's hot.} Warm paper-lantern light, steam curling through the frame, a rain-streaked window behind her, 50mm lens, shallow depth of field, fine film grain. Sound: bubbling broth, a sizzling pan, soft rain on the glass.
- Long single takes. Ask for one continuous shot with no cuts and give each time range one event. Our 15 second platform scene uses three 5 second beats, and every beat plays out in order in a single take.
- Characters that stay themselves. Name the details that must not change, like her face, red coat, scarf and suitcase, and Seedance 2.5 carries them from the first frame to the last.
- Dialogue in your language. Put spoken lines in curly braces and name the language before them. Our chef speaks her English line word for word in both the 480p preview and the 1080p clip.
- Sound made with the picture. Describe ambience and effects the way a sound designer would, like boots on snow, train brakes or rain on glass. On our long take the sound swells as the train pulls in.
- Preview, then finish in 1080p. A 5 second 480p preview is 15 credits, and upgrading that exact shot to 1080p is 80 credits, with the framing and motion you already saw.
- Frame it for the feed. The same prompt style works in 16:9, in 9:16 for vertical video and in 21:9 for widescreen, as our dancer and glacier clips show.
Seedance 2.5 vs Seedance 2.0: what the upgrade adds
Seedance 2.0 is also on Pixel Dojo. Seedance 2.5 builds on it with longer takes, flexible lengths, draft mode and a much bigger reference budget.
| Spec | Seedance 2.5 | Seedance 2.0 |
|---|---|---|
| Longest clip | 30 s | 15 s |
| Lengths you can pick | Any whole second from 4 to 30, or auto | 5, 10 or 15 s |
| Draft mode | Yes, preview at 480p, then upgrade the same shot to 1080p | No |
| Reference files | Up to 50: 30 images, 10 videos, 10 audio | Up to 15 in the separate reference tool: 9 images, 3 videos, 3 audio |
Seedance 2.5 pricing: credits per clip
Seedance 2.5 is included in every Pixel Dojo plan, with plans from $10/month. You pay in credits by the second of video, set by resolution: 3 credits per second at 480p, 6 at 720p and 16 at 1080p. Sound and dialogue come with every clip at no extra cost, and so does Thinking, an optional setting that plans long multi-beat prompts before it renders.
At the settings on this page, a 5 second clip is 15 credits at 480p, 30 credits at 720p and 80 credits at 1080p. Our 15 second single take at 480p was 45 credits, and a full 30 second clip at 1080p is 480 credits.
How to use Seedance 2.5 on Pixel Dojo
- 1
Write it as a shot note
Lead with shot size and camera move, then one specific subject, one action beat, light, lens and the sounds you want. For a long take, ask for one continuous shot with no cuts and split it into time ranges with one event each, the way our 15 second platform prompt does. Put spoken lines in curly braces and name the language before them. Original characters and invented products described in detail give the best results.
- 2
Preview it in draft mode
Turn on draft mode in the Seedance 2.5 tool to preview the shot at 480p. A 5 second preview is 15 credits and shows the exact framing and motion your 1080p clip will have.
- 3
Upgrade it to 1080p
Upgrade the preview to 1080p from its card and Seedance 2.5 renders that exact shot in full detail, 80 credits for 5 seconds. Previews stay upgradeable for 7 days in the tool and for 24 hours through the API.
Seedance 2.5 questions
What is Seedance 2.5?
Seedance 2.5 is ByteDance's AI video model for text to video, image to video from a first frame, and reference to video. It makes single takes from 4 to 30 seconds at 480p, 720p or 1080p in 16:9, 4:3, 1:1, 3:4, 9:16 or 21:9, with sound and dialogue generated alongside the picture. On Pixel Dojo you can use it in the browser or through the API.
What is Seedance 2.5 good at?
Long continuous takes with a consistent character, spoken dialogue and cinematic effects. Our 15 second platform scene plays three staged beats in one shot with no cuts, our chef speaks her English line word for word, and our marble lion bursts into golden sand in 1080p. It also handles action, fashion, aerial landscapes and 3D animation, as the examples above show.
How long can a Seedance 2.5 video be?
Any whole number of seconds from 4 to 30 in one request, twice the 15 second limit of Seedance 2.0. You can also pick auto, where the model chooses the length and you pay only for the seconds it makes.
Does Seedance 2.5 generate sound and dialogue?
Yes. Audio is on by default and made together with the picture: dialogue you write in curly braces, ambience, sound effects and music, with speech in 11 languages. Our chef clip speaks its English line word for word, and on our long take the sound swells as the train pulls in.
Can I guide Seedance 2.5 with reference images, video or audio?
Yes. One request can take up to 50 reference files: 30 images, 10 videos and 10 audio clips, with up to 30 seconds of reference video and 30 seconds of reference audio in total. You can also start from a first frame. Every clip on this page started from a text prompt alone. Reference video is billed by the second at the same rate as your clip.
How much does Seedance 2.5 cost?
Seedance 2.5 is included in every Pixel Dojo plan, with plans from $10/month, and uses credits by the second. A 5 second clip is 15 credits at 480p, 30 credits at 720p and 80 credits at 1080p, with sound included.
How do I use Seedance 2.5 on Pixel Dojo?
Open the Seedance 2.5 tool, write your shot, pick a length, resolution and aspect ratio, and generate. Developers can POST a prompt, duration, resolution and aspect_ratio to /api/v1/models/seedance-2-5/run and poll the job; API output links last one hour, so download your clips when the job finishes.
Make your first Seedance 2.5 clip
Write a prompt above or start from any example on this page. One subscription covers every model on Pixel Dojo.
Start with Seedance 2.5