PrivateGPT

PrivateGPT provides an API layer for building private AI applications with local model inference.

privategpt.dev
PrivateGPT homepage screenshot

About PrivateGPT

PrivateGPT is an API layer for building AI applications that run on local models. It provides building blocks such as a Messages API with streaming, file and artifact ingestion, retrieval with citations, and access to databases and CSV files. The API follows the Claude API model, so tools built for Claude-style or OpenAI-compatible interfaces can connect to it. The homepage also includes a built-in /ui workbench for testing, demos, and inspecting API requests.

The homepage states that PrivateGPT is designed to keep data inside the user's environment. This depends on the setup. Connecting MCP tools to online providers, or using external services such as n8n, can send data outside the local environment. The homepage describes PrivateGPT as the open-source application layer and points enterprise needs such as deployment, governance, user management, and auditability to Zylon, a separate product by the same team. Pricing is not stated on the homepage.

Key features

  • Messages API with streaming, async processing, and token counting
  • Upload and process documents and artifacts as context for AI workflows
  • Retrieval-augmented answers with source citations
  • Access to databases, CSVs, and tabular data through the API
  • Coordinates embeddings, retrieval, tools, and model calls behind one API
  • Custom tools and MCP connectors, run locally or with chosen online providers
Best for: Developers and teams building document-based AI applications that run on local models and need an API layer rather than a finished end-user product.
Good for
  • Provides a Messages API with streaming, async processing, and token counting
  • Supports local model servers including Ollama, llama.cpp, vLLM, and LocalAI
  • Works with tools that support a local OpenAI-compatible provider
Watch out for
  • Enterprise features such as RBAC, LDAP support, and SIEM audit logs are listed for Zylon, not PrivateGPT
  • Connecting MCP tools or workflows to online providers can send data outside the local environment
  • Pricing is not stated on the homepage

PrivateGPT at a glance

Type
Online tool
Pricing
Not confirmed. Check the site
Platforms
Web
Open source
Not stated
Language
English
Last checked
Oct 6, 2026 · Online when checked
Listed
Oct 5, 2026

PrivateGPT alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 12 alternatives

Cerebras Chat

chat.cerebras.ai

Cerebras Inference provides a service for running AI models from Cerebras.

Popular in AI Chatbots & Assistants

Well-known listings from the same topic.

All 238

PrivateGPT FAQ

Which local model servers does PrivateGPT support?
The homepage lists Ollama, llama.cpp, vLLM, LocalAI, and any preferred OpenAI-compatible inference server.
What is the difference between PrivateGPT and Zylon?
The homepage describes PrivateGPT as the open-source application API layer. It describes Zylon as enterprise AI infrastructure that adds deployment, governance, operations, user management, integrations, auditability, and support.
Can it be used without an internet connection?
A testimonial on the homepage states that apps built with PrivateGPT on your documents can work without an Internet connection. Whether this applies depends on the models and services used.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing