PrivateGPT is a unique project that enables the use of Large Language Models (LLMs) for document-based queries in a secure, offline environment. It features a FastAPI-based API that adheres to and expands upon the OpenAI API standard. This API offers both high-level and low-level functionalities for document ingestion, chat, completions, and retrieval of contextual chunks. The project also includes a Retrieval Augmented Generation (RAG) pipeline that utilizes LlamaIndex abstractions to separate the components and layers, thereby facilitating easy customization and extension. To assist users, PrivateGPT provides a comprehensive documentation site that covers all aspects from installation and configuration to usage and deployment. Additionally, it maintains a GitHub repository that encourages contributions and provides citation information, fostering a collaborative development environment.