LEANN
by StarTrail-org
LEANN is a vector database that lets you search your own files, emails, browser history and chat logs by meaning, using far less disk space than usual. It runs on your own laptop, so your data stays private and no cloud service is needed.
- Stars
- 13.5k +148 a day
- Licence
- MIT
- Runs on
- Your computer
- Setup
- Moderate
- Language
- Python
- Age
- 506 days
What you can do with it
- Index documents, emails, browser history and chat history, then search or ask questions about them on your own laptop
- Add it to Claude Code as a search service so the coding agent can find relevant code by meaning rather than keywords
- Use it as a small, portable vector database in your own Python apps built with LangChain or LlamaIndex
Make money with it
A freelancer could set up private, offline document search for small firms that cannot send data to cloud AI services, and charge for setup and support.
An idea, not financial advice. Check the licence before charging for anything built on it.
Ways a business could use it
- Build a private searchable knowledge base from company documents, emails and chat logs without paying for a hosted vector database
- Offer clients an offline question-and-answer system over their own files where data must not leave their machines
Can you use it commercially?
Free to use, change and sell, including in paid products. Keep the copyright notice.
Not legal advice. Read the project's licence file before building a product on it.
Getting it running
Install uv, clone the repository, then run uv pip install leann inside a virtual environment; building from source needs extra system libraries.
Follow the official instructions in the project README.
Things to check
- Some features, such as reading chat or email history, need access to sensitive personal data, so check permissions carefully
- Building from source can be fiddly, with extra system libraries and some Linux install quirks
- Answer generation needs a local model such as Ollama or an API key for a hosted one
Who's talking about it
Checked 28 Oct 2026. Summary drafted by AI from the project's README.
