Multimodal video generation with native audio, up to 20 seconds.

Two modes to choose from:

Generate mode: text to video, animate from a start frame, or first and last frame animation.

Keyframe mode:  add up to 6 reference images to guide a single video, paired with a prompt.

Duration: 5-20s.
Resolution: 720p or 1080p.
Audio generation: can be toggled on or off.

Access through the Create Video flow on Canvas, and open the Advanced Features to change modes and settings.