Seedance 2.5 — up to 30 seconds in one generation
The new Seedance generation is live in Clipia: text-to-video and image-to-video, clips up to 30 seconds with no stitching, up to 30 references and native audio.
We added Seedance 2.5, a new generation of the video model built for longer, coherent scenes. It creates up to 30 seconds in a single generation, so camera movement, characters and objects do not have to be stitched together from several short clips.
What changed
- Up to 30 seconds in one take: twice the maximum duration of Seedance 2.0.
- Two modes: generate from text or animate a source image.
- Up to 30 reference images: define the character, product, lighting, environment and style separately.
- Closer reference-video following: the model better preserves framing, intent and camera language.
- Native audio: sound can be generated together with the video.
Where it fits
Seedance 2.5 is a strong fit for ad scenes, Reels, product demos and short stories where continuous movement and subject consistency across camera angles matter.
How to try it
Open the video editor, choose Seedance 2.5 and set the duration, aspect ratio and references. Switch to Image to Video when you want to animate an existing image.
Try this model right now — no setup required.
Try Seedance 2.5
