# libre-chat

> Libre Chat is a self-hosted, open-source chatbot web service powered by local LLMs like Mixtral or Llama 2, with web UI, API, streaming responses, and document-based QA.

- **Category:** AI Chat
- **Licence:** MIT
- **Latest version:** 0.0.6
- **Open source:** oss
- **Languages:** Typescript, Jupyter Notebook, Python
- **Platforms:** Docker, Docker Compose, pip
- **Tags:** Openapi, Chatbot, Offline, LLM, Rag, Chatgpt, Gguf, Vector Search, Langchain, Docker, Python, Fastapi, Self Hosted, Llm inference, Large language models, Llama Cpp

## Description

Libre Chat is a free and open-source chatbot web service that lets you run large language models locally on your own hardware. It provides both a web UI and a REST/websocket API for interacting with models like Mixtral or Llama 2 in GGUF format via llama.cpp. The service supports generic conversation and experimental document-based question answering with vector search over uploaded documents (PDF, CSV, HTML, JSON, markdown). It runs fully offline with no API keys required, works on CPU (though slower), and can be deployed via Docker, docker-compose, or pip. Configuration is done through a YAML file. The project is in early development (pre-1.0) and the author recommends more mature alternatives like Ollama or GPT4All for production use.

## Links

- [Homepage](https://github.com/vemonet/libre-chat)
- [Website](https://vemonet.github.io/libre-chat)
- [Docker](ghcr.io/vemonet/libre-chat)
- [GitHub Repository](https://github.com/vemonet/libre-chat)
- [PyPI Package](https://pypi.org/project/libre-chat/)
- [Docker Image (GHCR)](https://github.com/vemonet/libre-chat/pkgs/container/libre-chat)
- [Issues](https://github.com/vemonet/libre-chat/issues)

---
Source: https://selfhost.directory/project/libre-chat