Seedance 2.0 AI Video Generator

Seedance 2.0 is a multimodal AI video generator. It takes text, images, video, and audio as input and returns a clip of 4 to 15 seconds, with the sound generated alongside the picture rather than added afterwards.

Three. Text to video works from a prompt alone. Image to video animates a still, or fills in the motion between a first and last frame. Reference to video takes images, clips, and audio together. Pick one per generation.

Seedance 2.0 takes up to 9 images, 3 video clips, and 3 audio tracks in one request, with video and audio each capped at 15 seconds in total. The images fix the look, the clips carry the camera move, the audio supplies the voice.

One prompt can describe several shots: a cut to a close-up, a change of angle, a move to another room. The character and the lighting hold across those cuts, so you get a small scene rather than three unrelated clips.

Yes, by default. Dialogue, effects, and ambience come out of the same pass as the picture, so a speaking character's mouth follows the track with no separate sync step. There is a switch in the panel if you would rather score it yourself.

Images as JPG, PNG, WebP, BMP, TIFF, or GIF up to 10MB. Reference video as MP4 or MOV up to 50MB. Reference audio as MP3 or WAV up to 15MB. In image mode you can paste an image URL instead of uploading.

Per second, scaling with model and resolution. The cheapest, Seedance 2.0 Mini, is 10 credits per second at 480p and 20 at 720p; Seedance 2.0 Fast is 17 and 34; Seedance 2.0 is 21, 42, and 105 for 480p, 720p, and 1080p. Five seconds at 480p on Mini comes to just 50. Reference video adds its own length to the billed seconds at a lower per-second rate. The exact figure sits on the generate button before you commit.

Usually a few minutes. Both Seedance 2.0 Mini and Fast return sooner than Seedance 2.0 — Mini being the cheapest and fastest, ideal for bulk drafts — and longer or higher-resolution clips take longer still. The page keeps checking on its own, so you can leave the tab open and come back to it.

Product and ad shots built from a single packshot, talking-head clips where the voice arrives with the picture, vertical cuts for TikTok, Instagram Reels, YouTube Shorts, and Facebook, anime and other stylised looks, and rough previz before an actual shoot.

Yes, but review every clip before it goes out. Check brand accuracy, likeness and copyright, and the rules of the platform you publish to. Treat Seedance 2.0 footage as a draft that needs a pass, not a finished asset.