Easy Apply BrushNet(applyBrushNet)

Description

Applies BrushNet inpainting conditioning to an EasyUse pipeline. The node takes a PIPE_LINE, an image, a mask, a BrushNet model selection, and several conditioning parameters, then returns the modified PIPE_LINE for continued processing.

This node belongs to the EasyUse/Inpaint category and is intended for use inside an EasyUse pipeline workflow.

Inputs

pipe

The EasyUse pipeline object to which BrushNet conditioning will be applied. The modified pipeline is returned as the output.

image

The input image to be inpainted.

mask

The inpaint mask. Areas covered by the mask are the regions BrushNet will inpaint or repaint.

brushnet

Selects which BrushNet model or configuration to use for the inpainting pass.

dtype

The computation precision used when running BrushNet.

Options:

• float16
• bfloat16
• float32
• float64

Lower-precision dtypes such as float16 or bfloat16 generally use less memory, while higher precision may improve numerical accuracy at the cost of speed and memory.

scale

Controls the strength of the BrushNet conditioning.

• Default: 1
• Minimum: 0
• Maximum: 10

Higher values make the inpainting follow the BrushNet conditioning more strongly. Use 0 to disable its influence.

start_at

The starting timestep at which BrushNet conditioning begins to be applied.

• Default: 0
• Minimum: 0
• Maximum: 10000

end_at

The ending timestep at which BrushNet conditioning stops being applied.

• Default: 10000
• Minimum: 0
• Maximum: 10000

With the defaults, BrushNet is applied across the entire denoising timestep range.

Outputs

PIPE_LINE

The updated EasyUse pipeline with BrushNet conditioning applied. This can be passed to downstream EasyUse nodes for further sampling or processing.

Usage Notes

• The image and mask should have matching dimensions, with the mask highlighting the region to be inpainted.
• To apply BrushNet only during part of the denoising process, adjust start_at and end_at. For example, setting start_at higher delays the BrushNet effect until later in sampling, while lowering end_at stops it earlier.
• The scale value lets you balance between following the BrushNet guidance and allowing the surrounding image content to influence the result.
• The returned PIPE_LINE is meant to be chained into the rest of the EasyUse workflow.