RecraftImageInpaintingNode - ComfyUI Built-in Node Documentation

Complete documentation for the RecraftImageInpaintingNode node in ComfyUI. Learn its inputs, outputs, parameters and usage.

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

This node modifies specific areas of an image based on a text prompt and a mask. It uses the Recraft API to intelligently edit only the masked regions while keeping the rest of the image unchanged.

Inputs

Parameter	Data Type	Required	Range	Description
image	IMAGE	Yes	-	The input image to be modified
mask	MASK	Yes	-	The mask defining which areas of the image should be modified
prompt	STRING	Yes	-	Prompt for the image generation (default: empty string, maximum length: 1000 characters)
n	INT	Yes	1-6	The number of images to generate (default: 1, minimum: 1, maximum: 6)
seed	INT	Yes	0-18446744073709551615	Seed to determine if node should re-run; actual results are nondeterministic regardless of seed (default: 0)
recraft_style	STYLEV3	No	-	Optional style parameter for the Recraft API. If not provided, defaults to “realistic_image” style
negative_prompt	STRING	No	-	An optional text description of undesired elements on an image (default: empty string)

Note: The image and mask must be provided together for the inpainting operation to work. The mask will be automatically resized to match the image dimensions. The prompt is validated and has a maximum length of 1000 characters. If a style_id from the Infinite Style Library is used, ensure it is not a Vector art style, as this may cause the API to return SVG data instead of an image.

Outputs

Output Name	Data Type	Description
image	IMAGE	The modified image(s) generated based on the prompt and mask. Returns one image per input image multiplied by the n parameter

---

Source fingerprint (SHA-256): c0871ae6510d42f5525701dd622f2f84a65103a7eec053635c4bd64102a2a4e6