Actions16
- LLM (Text Generation) Actions
- Image Generation Actions
- Video Generation Actions
- Text-to-Speech Actions
- Speech-to-Text Actions
- Music Generation Actions
- Embeddings Actions
- Content Moderation Actions
- Custom Inference Actions
Image Generation → Edit
AI-generatedSummary
Edit an input image by providing a text prompt describing the desired changes, optionally specifying the image source, output size, number of edited images, and advanced options like quality, style, seed, negative prompt, guidance scale, response format, and timeout.
Inputs
- Chute — Select a specific chute to use or enter a custom chute URL; the chute defines the image editing model and endpoint.
- Input Image — The image to edit, accepted as a URL, base64 data URI, or binary data from a previous node.
- Edit Prompt (required) — A textual description of the changes to apply to the input image.
- Size — The resolution of the output image, selectable from predefined options (256x256 up to 1792x1024).
- Number of Images — How many edited images to generate; multiple sequential API requests are made if more than one.
- Additional Options — Optional parameters including quality (standard, HD), style (natural, vivid), seed for reproducible results, negative prompt to specify undesired elements, guidance scale to control prompt adherence, response format (URL or base64), and maximum timeout in seconds.
Output shape
a list of edited images
Returns one or more edited images as separate items; each image can be returned as a URL or binary data depending on the response format chosen. If multiple images are requested, the node performs sequential API calls and outputs each edited image individually. Errors occur if input image is invalid or generation exceeds the specified timeout.
Examples
Example 1: Edit a single image by adding a red hat and brightening the scene, returning one 1024x1024 image in URL format using the 'https://image\.chutes\.ai' chute.
Set the chute to default or specify a custom chute URL, provide an image URL or binary data as input image, set the prompt to describe the edits (e.g. 'Add a red hat, make it brighter'), optionally adjust size to 1024x1024 and number of images to 1, and select desired output format and other options.