Practical overview
When it fits
Use Lanpaint when a ComfyUI workflow needs mask-constrained image editing with a pretrained model, without fine-tuning or backpropagation.
Who it is for
- ComfyUI users who want to add masked image editing to their workflows.
- Developers working with pretrained diffusion or rectified-flow models who need an inpainting sampler.
Documented uses
- Fill or replace masked image regions using text guidance and surrounding image context.
- Set up outpainting or generative-fill workflows in ComfyUI.
- Adjust sampler iterations and prompt-versus-image emphasis for local edits.
Requirements & limits
The repository is public and declares a GPL-3.0 license. The documented setup uses ComfyUI and ComfyUI-Manager or a custom_nodes installation.
- Lanpaint is an extension for ComfyUI, not a standalone image editor.
- The README recommends binary masks; opacity and smoothing are not preserved as soft mask values.
- “Think Mode” adds computation time; the README estimates five thinking iterations as five times slower than sampling without thinking.
- Prompt First mode emphasizes text guidance and is suggested for tasks such as character consistency, while the README notes a possible image-quality cost.
- The repository documents video inpainting workflows, but warns that longer sequences take significantly more processing time.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Install Lanpaint nodes in ComfyUI
Add the Lanpaint extension to an existing ComfyUI installation and load its nodes.
Before you start
- ComfyUI installed, or a ComfyUI version greater than 0.3.11.
- ComfyUI-Manager for the documented manager-based installation methods.
Steps
In ComfyUI-Manager, search for Lanpaint and install it.
Alternatively, use the manager’s Install via Git URL option with https://github.com/scraed/LanPaint.git, or clone the repository into ComfyUI/custom_nodes.
Restart ComfyUI.
Find the Lanpaint nodes in the sampling category.
Expected outcome
Lanpaint sampler nodes are loaded in ComfyUI for use in workflows.
Guide limitations
- This guide covers extension setup; it does not establish that a particular model or workflow will run successfully.
Run a documented image-inpainting example
Load a repository example workflow and queue an image-inpainting task in ComfyUI.
Before you start
- ComfyUI with Lanpaint nodes installed.
- The example’s workflow and image files.
- The model required by that example.
Steps
Download the files from an example folder in the repository.
Drag the example’s InPainted_Drag_Me_to_ComfyUI.png into ComfyUI to load its workflow.
Download the model specified by the example and load it in ComfyUI.
Place Masked_Load_Me_in_Loader.png in the workflow’s image-loading node for the inpainting mask, or use its Prepare Image node.
Queue the task.
Expected outcome
The workflow is set up to produce an inpainted result using the example’s model and masked input.
Guide limitations
- The required model varies by example.
- The project says Lanpaint requires binary masks; masks with smoothing or gradients are converted to binary.
- The README advises explicitly describing the desired masked content in the prompt.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 15 October 2026
- Original website title
- scraed/LanPaint · Official homepage title · retained in its source language
Related options
LanPaint-Diffusers — A separate Diffusers implementation identified by the Lanpaint repository.