explaingit

sanster/iopaint

23,107PythonAudience · vibe coderComplexity · 3/5StaleLicenseSetup · hard

TLDR

Free, open-source AI image editor that removes or replaces objects in photos by painting over them, no Photoshop skills needed.

Mindmap

mindmap
  root((IOPaint))
    What it does
      Remove unwanted objects
      Replace with text description
      Extend image edges
      Draw text into images
    How to use it
      Paint a mask over area
      AI fills it seamlessly
      Browser-based interface
      Command-line bulk processing
    Tech stack
      Python backend
      Multiple AI models
      GPU acceleration support
      Cross-platform
    Use cases
      Clean up photos
      Product image backgrounds
      Remove watermarks
      Erase blemishes
    Who benefits
      Photo enthusiasts
      Content creators
      Small businesses
      Budget-conscious users

Things people build with this

USE CASE 1

Remove unwanted people, objects, or watermarks from photos automatically.

USE CASE 2

Create clean product images by erasing backgrounds and replacing them with text descriptions.

USE CASE 3

Extend photos beyond their original edges or fill in missing areas seamlessly.

USE CASE 4

Batch-process multiple images via command line to clean up photo libraries.

Tech stack

PythonPyTorchNVIDIA CUDAApple MetalFastAPIReact

Getting it running

Difficulty · hard Time to first run · 1h+

Requires PyTorch with GPU support (CUDA or Metal) which has complex dependency chains and long compilation times.

Use freely for any purpose including commercial. Keep the notice and disclose changes to the patent grant.

In plain English

IOPaint is a free, open-source image editing tool that uses AI to remove or replace things in photos. The core feature is "inpainting", you paint over an area of an image with a brush (creating a "mask"), and the AI fills it in seamlessly, either removing the object entirely or replacing it with something new. The problem it solves is that removing unwanted elements from photos, a person in the background, a watermark, a blemish, or an object, used to require skilled manual editing in tools like Photoshop. IOPaint does this automatically using state-of-the-art AI models. You can erase objects so the background fills in naturally, replace them with different objects by typing a text description, draw text into images, or extend images beyond their original edges. You would use this if you want to clean up photos, remove distracting elements, create product images with clean backgrounds, or experiment with AI image editing without paying for cloud services. It runs entirely on your own computer, supporting CPU, NVIDIA GPU, and Apple Silicon, through a browser-based interface. It can also process images in bulk via the command line. The tool is written in Python and supports multiple AI models for different tasks.

Copy-paste prompts

Prompt 1
How do I set up IOPaint on my Mac with Apple Silicon to edit photos locally?
Prompt 2
Show me how to use IOPaint's inpainting feature to remove a person from the background of a photo.
Prompt 3
Can I use IOPaint to replace an object in an image by typing a description instead of painting?
Prompt 4
What AI models does IOPaint support, and how do I choose the best one for removing watermarks?
Prompt 5
How do I batch-process 100 product photos with IOPaint to remove their backgrounds?
Open on GitHub → Explain another repo

Generated 2026-05-18 · Model: sonnet-4-6 · Verify against the repo before relying on details.