Why choose it
- Native 4–30 second generation
- Animate, first/last, and image references
- Optional synchronized audio
Original model creator
ByteDance
Seedance 2.5
A next-generation text-and-image video model with native audio, 30-second shots, first/last frames, and deep reference control.
Loading Seedance 2.5 sample
Official sampleSourced output
Official sample · Results vary by prompt and settings
See the model move.
Direct answer
When should you use Seedance 2.5?
Seedance 2.5 is the Video Lab choice for longer directed shots and reference-heavy concepts. Generate up to 30 seconds from text, animate a starting image, define both ends of a transition, or guide identity and composition with up to 30 reference images while generating synchronized audio in the same pass.
Best for
Long-form ad concepts, cinematic one-takes, reference-led brand work, and scenes where sound drives action.
Technical readout
The supported shot envelope.
- Duration
- 4–30 seconds
- Resolution
- 480p or 720p
- Aspect ratio
- Landscape, portrait, square, and ultrawide formats
- Audio
- Optional synchronized audio
What to consider
- Video Lab currently exposes 480p and 720p output
- Video and audio reference uploads are not yet exposed in the Clip Studio reference workflow
Prompt direction
How to prompt Seedance 2.5
For a longer shot, write the action as ordered visual beats and name each reference explicitly—for example, @Image1 for the subject and @Image2 for the product.
Start a Seedance 2.5 promptQuestions, answered
Seedance 2.5 FAQ
What is Seedance 2.5?
Seedance 2.5 is ByteDance’s newer native-audio video model for text, image animation, start/end frames, and multimodal reference direction. Clip Studio exposes its text and image workflows through Video Lab.
How long can a Seedance 2.5 video be?
Video Lab supports fixed Seedance 2.5 durations from 4 to 30 seconds in a single generation.
Can Seedance 2.5 use reference images?
Yes. Clip Studio Video Lab supports up to 30 Seedance 2.5 reference images, alongside single-image animation and first-to-last-frame direction.
What is the difference between Seedance 2.5 and Seedance 2.0?
Seedance 2.5 extends native generation to 30 seconds and accepts a much larger reference set. Seedance 2.0 remains useful when you need its supported image modes at resolutions above 720p.
Change the lens
Compare Seedance 2.5 with another model.
Independent testByteDanceOriginal model creatorSeedance 2.0
A flexible text-and-image model with first/last-frame direction, up to nine references, broad formats, and optional audio.See model
Official sample
GoogleOriginal model creatorVeo 3.1
A premium text-and-image video model for cinematic shots, guided frames, references, optional audio, and up to 4K output.See model
Official sample
KuaishouOriginal model creatorKling 3 Pro
A controllable text-and-image model for expressive movement, first/last frames, subject references, and optional audio.See modelByteDance lens selected
Direct your first Seedance 2.5 shot.
Write the prompt, choose supported controls, and see the exact credit quote before you generate.
Open Seedance 2.5 in Video Lab