Skip to main content
Edit Video modifies an existing video according to text instructions. Depending on the selected model, you can supply reference images and a first frame to guide the result.

​
When to use

  • Style changes
  • Object edits
  • Iterating on footage
Common patterns: upload_video → edit_video, or generate_video → edit_video.

​
Inputs

​
video
required
The video to edit. Required. Supports loop mode.
​
text
required
Text describing the edit to apply. Required. Supports loop mode.
Inline instructions. You don’t need a separate Text Prompt node for fixed instructions — type them directly on the node. The inline field appears whenever the instructions input is unconnected and disappears once you wire a text input in. Connect a Text Prompt node only when you want to drive the instructions from upstream or expose them as an app input. Type @ to mention a connected input by name — text inputs insert their content into the instructions, media inputs insert their name.
​
image
Reference images to guide the edit. Available only when the selected model supports reference images. Accepts multiple.
​
image
An image used as the opening frame. Available only when the selected model supports reference images.

​
Outputs

​
video
The edited video.
This node runs on a selectable model and consumes credits. Model-specific options (such as aspect ratio, resolution, and audio handling) are available depending on the model you choose. Pick the model in the node’s config panel. See Video models for the options, and Model settings for what each setting does.