AnythingLLM

AnythingLLM is a comprehensive desktop application available for Mac, Windows, and Linux, designed to let you download and locally run any LLM, enabling private chats with your documents. By default, the LLM, embeddings, and even the vector database are stored locally and can operate fully offline.

To utilize more robust models, embedders, or vector databases, you can link to over 20 integrated providers to enhance your AnythingLLM setup.

With Ollama pre-installed, a complete RAG chatbot is ready for use with no setup required and complete privacy. Upload various file types like PDFs, MP3s, and text files.

AnythingLLM also includes a complete developer API for custom integrations, along with multiple data source connectors, such as GitHub and YouTube.

AnythingLLM is an MIT-licensed, fully open-source project that you can modify.

For a multi-user setup, AnythingLLM provides a self-hosted Docker image supporting concurrent users. AnythingLLM is designed to be the ultimate RAG-based chatbot solution.