Seedance 2.5 Sharper motion, richer native audio

Seedance 2.5 is the newest generation of ByteDance's Seedance video model. It turns a text prompt or a single reference image into cinema-quality video with native audio, holding motion together more steadily and following prompts more closely than earlier Seedance releases.

Seedance 2.5 is the top-quality tier: it keeps motion and character detail consistent across a shot and handles longer, more specific prompts. Seedance 2.0, Fast and Mini cost fewer credits per second, so they remain a good fit for drafts and quick iterations.

Text-to-video and image-to-video. Write a prompt on its own, or upload a reference image as the first frame and describe how the scene should move from there.

Yes. Audio is generated together with the video by default — dialogue, sound effects and ambience synced to what is on screen. Switch Audio off in the form whenever you need a silent clip.

Seedance 2.5 supports 480p and 720p, any duration from 4 to 15 seconds, and 16:9, 4:3, 1:1, 3:4, 9:16 and 21:9 aspect ratios. Adaptive is available too, letting the model keep the framing of your reference image.

Credits are billed per second of video, and 720p costs more per second than 480p. The exact total for your current settings appears on the Generate button before you submit, and credits are refunded automatically if a generation fails.

Yes. Videos you generate with Seedance 2.5 on Nano Banana can be used for both personal and commercial projects, including social media and marketing.