zylon-ai/private-gpt

Health v2 · maintenance only

Adoption not part of the score

57558 stars · 7622 forks observed · 2026-10-05

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-21, confidence 0.97

PrivateGPT is an open‑source API framework that turns locally hosted LLMs into production‑grade AI services. It provides a Claude‑style API, RAG, tool execution and database connectors while keeping all data on‑premise.

Use cases

• Build a private ChatGPT‑like assistant that runs on a local LLM without sending data to the cloud.
• Add retrieval‑augmented generation (RAG) with citations to an internal knowledge base.
• Expose custom tools such as web search, code execution, or SQL queries through an OpenAI‑compatible endpoint.
• Create enterprise AI applications that need strict data sovereignty and on‑prem deployment.
• Develop AI‑powered workflows that combine LLM responses with database or CSV lookups.

When to choose

• You need an OpenAI‑compatible API for a self‑hosted LLM and want built‑in RAG, tool support, and MCP connectors.
• Your organization requires full data privacy and on‑premise deployment of AI services.
• You want a ready‑made framework to accelerate building private AI products without writing low‑level inference code.

When to avoid

• You are looking for a cloud‑hosted solution that manages model hosting for you.
• Your project only needs a simple LLM inference wrapper without RAG or tool orchestration.
• You require support for non‑OpenAI‑compatible inference servers that do not expose `/v1/chat/completions`.

Facets

framework · maturity active

api-framework agent-framework http-server rag developer-tools database-driver artificial-intelligence large-language-models chatbots backend python windows cli openai-compatible on-prem private-ai vector-database tool-integration mcp text-to-sql retrieval-augmented-generation docker linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="zylon-ai/private-gpt")