Back to events
Model releaseCritical global significanceConfirmed confidence

ByteDance Seed releases Seedance 2.0

Seedance 2.0 uses a unified multimodal architecture for joint audio-video generation, available through Jimeng, Doubao, and Volcano Ark.

Event details

Seedance 2.0 accepts text, image, audio, and video inputs, and can reference multiple assets at once to generate, edit, and extend footage, producing multi-shot audio-video clips up to 15 seconds. The model improves complex motion, physical accuracy, subject consistency, camera movement, and stereo audio.

Why it matters

Seedance 2.0 brings multi-asset referencing, video editing, and native joint audio-video generation into shipping products, moving video models toward complete creative workflows.

93/100Global significance score. Regional effects are recorded only when the evidence supports a meaningful difference.

Access notes

ByteDance Seed lists access through Jimeng, Doubao, and Volcano Ark.