Seedance 2.5

Seedance 2.5 is a multimodal model that takes text, images, video and audio together and generates a cinematic video. A single take can run up to 30 seconds with native synced audio, prompt-driven camera moves and output up to 1080p. You can lock subject and style with those references, then restyle, swap an element or extend a shot in plain language.

Yes. New accounts get free credits, which is enough to generate your first videos and judge the quality yourself. After that you can buy a credit pack or a subscription on the pricing page.

A single generation can run from 4 to 30 seconds as one continuous shot. Longer pieces are usually built by generating several shots and cutting them together, or by extending an existing clip through video editing.

You can start from a text prompt alone, or add images, video and audio as references in Seadance 2.5. Images work as a start frame or a subject and style guide, video guides motion, and audio can drive timing. You can combine up to 50 references in one generation.

Seedance 2.5 outputs 480p, 720p or 1080p. You can set the frame to 16:9, 9:16, 4:3, 3:4, 1:1 or 21:9, or leave it on Auto. Start from an image and the output follows that first frame's ratio.

Videos you generate with Seedance 2.5 on a paid plan can be used in commercial work, including ads and client deliverables. Check the terms and the AIGC acceptable use policy for the details and the content rules that apply.

Yes. Upload the clip, describe the change in plain language, and Seadance 2.5 applies it while keeping the rest of the frame intact. It handles restyling, replacing an element and extending a take.

Direct the shot rather than describing a mood. Name the subject, the action, the lens or camera move and the lighting, keep it to one clear beat per generation, and add a reference image whenever the look matters.