BriaGenFill - ComfyUI Built-in Node Documentation - ComfyUI

This node generates objects or scenery inside a masked region of an image using the Bria API. It uploads the image and mask, sends the prompt to the Bria generative fill service, waits for the operation to complete, and returns the edited image. This is a paid API operation (US$0.0429 per request).

Inputs

Parameter	Description	Data Type	Required	Range
`image`	The input image to edit.	IMAGE	Yes	-
`mask`	White areas are filled with generated content, black areas are preserved. The mask is binarized before sending, so partially painted areas count as white. Must have the same aspect ratio as the image.	MASK	Yes	-
`prompt`	Description of what to generate inside the masked region. Must contain at least 1 character.	STRING	Yes	-
`negative_prompt`	A prompt describing content to avoid in the generated result. If left empty, it is not sent to the API.	STRING	Yes	-
`refine_prompt`	Automatically adjust the prompt for better results; disable to use the prompt exactly as written. (default: true)	BOOLEAN	Yes	true false
`seed`	Seed for the generation process. (default: 42)	INT	Yes	1 to 2147483647
`moderation`	Moderation settings for the request. When set to “true”, the nested moderation options described below are applied. (default: “false”)	COMBO	Yes	”false" "true”

Note: The prompt must not be empty, and the mask must have the same aspect ratio as the image. When moderation is set to “true”, the following nested boolean options are available:

prompt_content_moderation (default: false): Applies content moderation to the prompt.
visual_input_moderation (default: false): Applies content moderation to the input image.
visual_output_moderation (default: false): Applies content moderation to the output image.

Outputs

Output Name	Description	Data Type
`image`	The resulting image with the masked region filled by the generated content.	IMAGE

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

Source fingerprint (SHA-256): 0d9babfa5e14c03f73d2b5befbd1c5cd1f5ffc685a0d7ccb3db09cfec51ba4fa