Canvas

Video Node

The Video node converts a generated image into a short-form video clip using image-to-video AI models. Perfect for creating ad content, social media reels, and animated product showcases.

How to Use

  1. Add a Video node from the Nodes panel.
  2. Connect an Image node's output to the Video node's input.
  3. Choose a video model from the dropdown.
  4. Optionally add a motion prompt describing the desired movement.
  5. Click Generate.

Fields

  • Model - Select from available video models (Kling, Veo, Hailuo, Wan).
  • Prompt - Describe the motion or action for the video (e.g. "character turns and smiles at the camera").

Connections

A Video node accepts input from:

  • Image Node - the generated image becomes the first frame of the video.

Available Models

  • Kling 3.0 v3 Pro - Character-aware video with @Element tags
  • Kling O3 - Kuaishou's newest flagship, 3-15s with native audio
  • Veo 3 - Google's premium video with native audio
  • Sora 2 - OpenAI's exceptional video generation

Motion Transfer Models

Connect a reference video to re-render your AI character with the motion from that video.

  • Kling 3.0 v3 Pro Motion - Best quality, up to 30s, supports @Element facial consistency
  • Kling 3.0 v3 Std Motion - Fast v3 motion transfer, up to 30s
  • DreamActor V2 - Multi-character and non-human motion transfer, up to 30s

Tips

  • Generate the best possible image first, then animate it - video quality depends heavily on the source image.
  • Keep motion prompts simple: "character smiles", "slow zoom in", "product rotation".
  • Video generation takes longer than images (30-120 seconds depending on the model).
Was this helpful?
Image NodeConnections