Seedance 2.0 Official Launch

ByteDance launched Seedance 2.0, a video generation model with multimodal input and improved controllability, on its official blog.

Supports up to 9 images, 3 video clips, 3 audio clips, and text as input
Generates 15-second multi-shot audio-video output with dual-channel audio
Improves physical accuracy, motion stability, and instruction following compared to version 1.5
Available on the Seedance 2.0 homepage at seed.bytedance.com/seedance2_0

Video creators and AI developers now have access to a model with enhanced multimodal input and control, potentially improving production workflows.

Sources