PrivateGPT

PrivateGPT

by imartinez

Ask questions to your private documents locally using LLMs — no internet required

Open Source Edge Computing API macOS Windows Linux Docker
Visit Product
190 upvotes 4,343 views

About

PrivateGPT is an open-source project that lets you interact with your documents using large language models in a completely private environment — no data ever leaves your computer and no internet connection is required. It answers questions about your PDFs, Word documents, text files, and other documents by ingesting them locally and using an LLM running on your own hardware.

The project uses Llama models (via Ollama or llama.cpp), local embeddings, and a local vector database (Chroma) to create a fully self-contained document Q&A system. This makes it ideal for lawyers, medical professionals, journalists, researchers, and anyone working with sensitive documents who cannot use cloud-based AI services for confidentiality or regulatory reasons.

PrivateGPT has become one of the most starred repositories on GitHub in the AI space, reflecting the huge demand for privacy-first AI tools. It can be used as a standalone application or integrated into larger private AI deployments.

Product Features

- 100% local: no internet, no cloud, no data sharing
- Document ingestion: PDF, Word, TXT, CSV, HTML, and more
- Local LLM inference via Ollama or llama.cpp
- Local embeddings for document vectorization
- Chroma local vector database for document storage
- REST API for integration with other applications
- Web UI for easy interaction
- Streaming responses for real-time output
- Source citation: shows which document passages were used
- Docker support for easy deployment

About the Publisher

PrivateGPT was created by Iván Martínez and released as open-source in May 2023. The project exploded in popularity, gaining over 50,000 GitHub stars within its first month — one of the fastest-growing open-source repositories in history. It has been maintained and significantly improved by the community, with enterprise deployments built on top of the codebase by companies seeking private AI capabilities.