Run a fast RAG app with 97% storage savings. For founders, backed by 12k+ GitHub stars
11,850 stars1,054 forksPythonUpdated 6/2/2026100% free ยท open source
What it does
LEANN allows you to run a fast and accurate Retrieval-Augmented Generation (RAG) application with significant storage savings, making it possible to deploy AI models on personal devices while maintaining privacy
When to use it
โขWhen you need to deploy AI models on devices with limited storage capacity
โขWhen you require a high level of data privacy for your AI application
โขWhen you want to reduce the latency and improve the responsiveness of your RAG application
Quick start
1Install LEANN from the GitHub repository: https://github.com/StarTrail-org/LEANN
2Set up your Python environment to run LEANN, ensuring you have the required dependencies
3Prepare your dataset and configure LEANN to run your RAG application
4Test and fine-tune your LEANN-powered RAG application for optimal performance