Create a video node
Generate animated video content from a text description or a starting image.
What Create Video produces
Create Video generates short video clips using AI video models. The result is placed on the canvas as a video layer that you can reposition and integrate into your design.
Two modes: text-to-video and image-to-video
| Mode | Starting point | When to use |
|---|---|---|
| Text-to-video | A text description only | You want to generate a scene or motion from scratch |
| Image-to-video | An image + optional text | You want to animate an existing image or canvas layer |
The models available in each mode differ — check the model selector to see which support each input type.
Generating a video
- Open the prompt input.
- Select the Create Video mode.
- Choose a video model from the model selector.
- Write your prompt describing the motion, scene, and mood:
- Include camera movement: "slow pan left", "dolly zoom", "static shot"
- Include subject motion: "a flag waving", "water rippling", "person walking"
- For image-to-video, attach a starting image or select a canvas layer as a reference.
- Configure available parameters such as duration and aspect ratio.
- Click Generate.
Video generation takes longer than image generation — typically 30 seconds to a few minutes depending on the model.
Prompt tips
| Goal | Example prompt |
|---|---|
| Cinematic scene | "A coastal cliff at sunset, waves crashing below, slow upward camera tilt, golden hour lighting" |
| Product animation | "A perfume bottle rotating slowly on a white surface, soft studio lighting, gentle sparkle effect" |
| Abstract motion | "Abstract liquid color flow, deep blues and purples, looping motion" |
| Animate an image | Attach the image + "animate this gently, subtle breeze through the trees" |
After generation
- The video layer appears on the canvas.
- Use the video layer controls to mute, loop, replace, or export the clip.