Live model choices
- Seedance 2.5 and Seedance 2.0
- Seedance 1.5 Pro
- Kling 3.0 Pro
- WAN 2.6 and WAN 2.5
- Motion and digital-human models for supported inputs
Still image to motion
Upload a source image, describe the intended motion, and choose from the image-to-video models actually enabled in the Cdance editor.
Depending on the selected model, Cdance exposes 3–30 second durations, up to 4K output, native sound, camera lock, multi-shot, or motion-reference controls.
Use a sharp image with a clearly visible subject and enough space in the direction of motion. Describe movement rather than repeating every visual detail already present in the image.