LumaImageEditNode2 - ComfyUI Built-in Node Documentation - ComfyUI

This node edits an existing image using a text prompt, powered by the Luma UNI-1 model. It takes a source image and a description of the desired change, then generates a new edited version of the image. You can choose between the uni-1 and uni-1-max models, adjust the style, enable web search, and optionally provide up to 8 reference images.

Inputs

Common Inputs

Parameter	Description	Data Type	Required	Range
`model`	Model to use for editing. Selecting a model reveals the model-specific options below.	DYNAMIC_COMBO	Yes	`"uni-1"` `"uni-1-max"`
`source`	Source image to edit.	IMAGE	Yes	-
`prompt`	Description of the desired edit. 1–6000 characters. Default: "" (empty string; the request is invalid until at least one character is entered).	STRING	Yes	1 to 6000 characters
`seed`	Seed controls whether the node should re-run; results are non-deterministic regardless of seed. Default: 0.	INT	Yes	0 to 2147483647

uni-1 and uni-1-max Inputs

These options are shared by both the uni-1 and uni-1-max models.

Parameter	Description	Data Type	Required	Range
`style`	Style preset. `"auto"` picks based on the prompt; `"manga"` applies a manga/anime aesthetic and requires a portrait aspect ratio (2:3, 9:16, 1:2, 1:3). Default: `"auto"`.	COMBO	Yes	`"auto"` `"manga"`
`web_search`	Search the web for visual references before generating. Default: false.	BOOLEAN	Yes	`true` `false`

Reference Inputs

Parameter	Description	Data Type	Required	Range
`image_ref`	Growable slot: connect up to 8 reference images (`image_1` through `image_8`) for style/content guidance. Optional.	IMAGE	No	0 to 8 images

Notes:

The prompt must be between 1 and 6000 characters.
The style, web_search, and image_ref inputs appear when model is set to either "uni-1" or "uni-1-max".
Both models support the same model-specific options, including up to 8 reference images.
The "manga" style requires a portrait aspect ratio (2:3, 9:16, 1:2, or 1:3).
Connecting more than 8 reference images raises an error.

Outputs

Output Name	Description	Data Type
`image`	The edited image generated by the Luma UNI-1 model, returned in PNG format.	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): 66f62bb2807759edb405c2caeeefe32c341920924e267c32449a620190b9a7ab