Render one continuous 30-second take with up to 50 multimodal references — local region editing re-cuts a finished ad for any market.
Seedance 2.5 is the newest generation of the Seedance AI video model, built to create longer, higher-quality cinematic videos from text prompts, images, audio, video clips, and multiple reference assets. Compared with previous versions, it introduces native 30-second video generation, improved character consistency, and significantly expanded multimodal reference support, making it suitable for professional storytelling and commercial video production.
Have feedback for the maker?
Sign in to leave a review, report a bug, or suggest a feature.
Kamrul Agarwal
