comfyui-inpaint-nodes
ComfyUI Inpaint Nodes
Nodes for better inpainting with ComfyUI: Fooocus inpaint model for SDXL, LaMa, MAT,
and various other tools for pre-filling inpaint & outpaint areas.
Fooocus Inpaint
Adds two nodes which allow using Fooocus inpaint model.
It’s a small and flexible patch which can be applied to your _SDXL checkpoints_ and
will transform them into an inpaint model. This model can then be used like other inpaint
models to seamlessly fill and expand areas in an image.
Download models from lllyasviel/fooocus_inpaint and place them into ComfyUI/models/inpaint.
[!NOTE]
Make sure to use the regular version of a checkpoint to create an inpaint model – distilled merges (Turbo, Lightning, Hyper) do not work.
Inpaint Conditioning
Fooocus inpaint can be used with ComfyUI’s _VAE Encode (for Inpainting)_ directly. However this does
not allow existing content in the masked area, denoise strength must be 1.0.
_InpaintModelConditioning_ can be used to combine inpaint models with existing content. The resulting
latent can however _not_ be used directly to patch the model using _Apply Fooocus Inpaint_. This repository
adds a new node VAE Encode & Inpaint Conditioning which provides two outputs: latent_inpaint (connect
this to _Apply Fooocus Inpaint_) and latent_samples (connect this to _KSampler_).
It’s the same as using both _VAE Encode (for Inpainting)_ and _InpaintModelConditioning_, but less overhead
because it avoids VAE-encoding the image twice. Example workflow
Inpaint Pre-processing
Several nodes are available to fill the masked area prior to inpainting. They avoid seams as long as the
input mask is large enough.
Expand Mask
Expands (grow) a binary mask by a certain number of pixels, and optionally blurs (feather) the mask for a smoother transition at the edges.
Shrink Mask
Shrinks (erodes) a binary mask by a certain number of pixels, and optionally blurs (feather) the mask for a smoother transition at the edges.
Stabilize Mask
Maps mask values very close to 1 to be exactly 1.0. This helps avoid numerical
issues introduced by previous operations. ComfyUI in various steps only
considers noise masks to be opaque when they are >= 1.0 (eg. via round), and
values like 0.9999… can lead to masks not working at all.
Fill Masked
This fills the masked area, with a smooth transition at the border. It has 3 modes:
neutral: fills with grey, good for adding entirely new contenttelea: fills with colors from surrounding border (based on algorithm by Alexandru Telea)navier-stokes: fills with colors from surrounding border (based on fluid dynamics described by Navier-Stokes)| Input | Neutral | Telea | Navier-Stokes |
|-|-|-|-|
| | | |
Blur Masked
This blurs the image into the masked area. The blur is less strong at the borders of the mask.
Good for keeping the general colors the same.
| Input | Blur radius 17 | Blur radius 65 |
|-|-|-|
| | | |
Inpaint Models (LaMA, MAT)
This runs a small, fast inpaint model on the masked area. Models can be loaded with Load Inpaint Model
and are applied with the Inpaint (using Model) node. This works well for outpainting or object removal.
The following inpaint models are supported, place them in ComfyUI/models/inpaint:
| Input | LaMa | MAT |
|-|-|-|
| | | |
Inpaint Post-processing
Color Match (Masked)
Can mitigate color/brightness shifts when inpainting. Requires the original
input image as reference, and the denoising output as target. This node is
useful only if the color shift occurs in the entire output, even the parts that
were excluded with a _noise mask_. In this case you can provide the same mask to
the Color match node as exclude mask. Colors will be anaylised based on the
changes _outside_ the masked area. Then the correction is applied to the entire
target image.
Particularly useful for refinements or edits done with Flux 2 Klein.
Denoise to Compositing Mask
Takes a _mask_, an _offset_ (default 0.1) and a _threshold_ (default 0.2).
Maps mask values in the range of \[_offset_ → _threshold_\] to \[0 → 1\].
Values below offset are clamped to 0, values above threshold to 1.
This is particularly useful in combination with ComfyUI’s “Differential Diffusion” node, which allows to use a mask as per-pixel denoise strength.
Using the same mask for compositing (alpha blending) defeats the purpose, but no blending at all degrades quality in regions with zero or very low strength. This node creates a mask suitable for blending from the denoise-mask.
Example Workflows
Example workflows can be found in workflows.
Installation
Use ComfyUI Manager and search for “ComfyUI Inpaint Nodes”.
_or_ download the repository and put the folder into ComfyUI/custom_nodes.
_or_ use GIT:
cd ComfyUI/custom_nodes
git clone https://github.com/Acly/comfyui-inpaint-nodes.gitRestart ComfyUI after installing!
OpenCV is required for _telea_ and _navier-stokes_ fill mode:
pip install opencv-python