Seedance 2.5 is ByteDance's multimodal AI video model and the successor to Seedance 2.0. It generates up to 30-second cinematic clips at native 4K with synchronized audio from text, images, audio, video, and 3D references.
Key features:
- Text-to-video, image-to-video, and video-to-video workflows
- Up to 50 multimodal references in a single joint input
- Native 4K output with audio generation
- Director-level control over camera, lighting, motion, and physics
- Multi-shot consistency for characters, wardrobe, and scenes
- Free credits to try online, with API access for production workflows
