Seeddance is a unified AI creative platform that combines multiple video, image and music models into a single multimodal workflow. Users can upload up to 12 assets (images, videos, audio, text), use a universal @ reference syntax to assign roles to each asset, and generate 4–15 second cinematic videos with built-in dual-channel stereo audio, lip-sync, camera replication, visual consistency across frames, and editing/extension features. It includes a credit-based pricing model, daily free credits, and plans for creators and teams.

Seedance 2.0 is an advanced AI video generation platform that converts text, images (and multimodal inputs) into coherent 4–15 second cinematic videos at up to 2K resolution and 24 FPS. It introduces Acoustic Physics Fields to generate environment-aware audio, World ID to preserve character identity across shots, and a World-MMDiT architecture that models gravity and motion for realistic physics. The system composes multi-shot narratives, syncs audio natively, and provides API access and node-based controls for professional creative workflows.