Lama Cleaner: An Advanced, Self-Hostable Inpainting Tool Powered by AI

Lama Cleaner is a cutting-edge, free, and open-source inpainting tool that harnesses the power of state-of-the-art AI models. This versatile tool empowers users to remove unwanted objects, defects, or even people from their pictures effortlessly. Additionally, it allows users to erase, replace, or modify any element within their images, offering exceptional creative possibilities.

Features of Lama Cleaner:

• Free and Open-Source: Lama Cleaner is entirely free to use and is open-source, meaning users have access to the underlying code and can customize it as needed.
• Fully Self-Hostable: Users can deploy and run Lama Cleaner on their own servers, ensuring complete control and privacy over their data.
• CPU, GPU & M1/2 Support: Lama Cleaner is designed to support various hardware setups, including CPU, GPU, and M1/2 processors, maximizing performance for different configurations.
• Windows 1-Click Installer: The tool offers a convenient 1-click installer for Windows users, simplifying the setup process.
• Native macOS App (Work in Progress): The team behind Lama Cleaner is actively working on a native macOS application to provide a seamless experience for macOS users.
• Multiple State-of-the-Art AI Models: Lama Cleaner leverages several advanced AI models, such as LaMa, LDM, ZITS, MAT, FcF, and Manga, enabling users to achieve precise inpainting and modifications.
• Erase and Replace Models: The tool offers both erase and replace models, including Stable Diffusion and Paint by Example, offering versatility in image editing.
• Post-Processing Plugins: Lama Cleaner supports various post-processing plugins, such as RemoveBG for background removal, RealESRGAN for super-resolution, GFPGAN for face restoration, RestoreFormer for face restoration, and Segment Anything for interactive object segmentation.

Quick Start Guide:

Lama Cleaner streamlines the usage of state-of-the-art AI models with just two simple commands:

# In order to use the GPU, install the cuda version of pytorch first.
pip install torch==1.13.1+cu117 torchvision==0.14.1 --extra-index-url https://download.pytorch.org/whl/cu117
pip install lama-cleaner
lama-cleaner --model=lama --device=cpu --port=8080

With these commands, Lama Cleaner will be up and running at http://localhost:8080, ready to serve your inpainting and image editing needs.

For more command line arguments and detailed documentation, visit the Lama Cleaner Documentation.

Development:

If you plan to modify the frontend and recompile it yourself, follow these steps:

Frontend:

• Install dependencies: Navigate to the lama_cleaner/app/ directory and run pnpm install to install the necessary frontend dependencies.
• Start development server: Use pnpm start to launch the development server and experience the frontend changes in real-time.
• Build: To compile the frontend, execute pnpm build.

Discover the immense potential of Lama Cleaner and experience the convenience of AI-powered inpainting and image editing. As an open-source tool, Lama Cleaner invites contributions and customization, fostering a collaborative and innovative community of users and developers.

Note: The responsible and ethical use of Lama Cleaner is vital to ensure compliance with relevant policies and regulations. It should only be used for authorized purposes and not for any malicious or unauthorized activities.

DOWNLOAD TOOLS