Seedance 2.0 is ByteDance's advanced multimodal AI video generation model lets you create videos with synchronized audio from text, image, video, and audio. Use up to 15 references for consistent characters, scenery and audio across shots.
Combine up to 9 images, 3 videos and 3 audio clips per project. Seedance 2.0 learns effects, camera moves, actions and editing styles — and replicates complex or popular scenes with a single click while keeping characters consistent.
Reproduce character details, composition and sound from your references. Unify font styles and precisely control pace and rhythm so scene transitions feel natural — the full creative process stays controllable and convenient.
Generate new storylines or continue existing videos with natural plot and shot connection. Audio-visual sync stays tight in single- and multi-person scenes — narration, environmental SFX and visuals locked together for real cinematic feel.
From concept to final cut in one place — storyboard with image models, plan shots and scenes, bring characters to life with lifelike digital humans. Story, visuals and scenes work together as a complete one-stop solution.
One of ByteDance's flagship video models, wired up here with multimodal reference input, native audio and cinematic camera motion.
A text prompt plus up to 9 reference images, 3 reference videos and 3 reference audio clips per generation.
Between 4 and 15 seconds. Leaving Duration on Auto lets the model pick, and is charged at the 15-second ceiling for the resolution you picked.
Switch models from the panel — Seedance 2.0 Mini runs the same engine faster and cheaper.
