Generate a video clip
Generate a video clip from a text prompt, an image, or an existing video on the board.
Before you start
- An image or video on the board (optional — you can also generate from text only).
Steps
- Select an image or video as a reference (optional).
- Use the Generate Video Clip quick action, or describe what you want in the agent input.
- Pick a video model with
@(see Video models). - Choose a duration — Short (4s), Medium (6s), or Long (10s), depending on the model.
- (Optional) Add start or end frames for more control (see Use start and end frames).
- Describe the scene. The agent generates the clip on the board.
Tips
- Video generations typically take 1–8 minutes depending on the model.
- You can queue up to 5 generations per session.