GPT Researcher

by GPT Researcher (open source)

Open source research agent that writes cited reports

Reviewed by the the30top research team · Last reviewed 2026-10-27

Overview

GPT Researcher researches the web and your own documents, then writes a detailed report with citations. It runs as a Python package, a FastAPI server or a Docker image, works with 20+ retrievers and any major LLM, and exposes an MCP server. It is free to use, but you supply and pay for your own LLM and search API keys, and you carry the setup and maintenance work yourself.

Our verdict

The best route to a customizable research agent if you have engineers to run it. Non-technical teams will be better served by a hosted product.

Best for

Developers and technical teams who want a self-hosted research agent they fully control.

Avoid if

You want a managed product with support, a UI for non-technical users and compliance paperwork.

Pros

Free, open source and self-hostable
Works with any major LLM and many search engines
Can combine local documents with web research

Cons

You maintain and host it yourself
You pay your own LLM and search API costs
No vendor support or compliance documentation

Key capabilities

Parallel researchers that read sources at the same time20+ retrievers including Tavily, Google, Bing, Exa, arXiv and PubMedDeep research that explores a tree of subtopicsHybrid research over local documents and the webExport to PDF, Word, Markdown, JSON and CSVMCP server for use from Claude, Cursor and other clients

Frequently asked questions

Is GPT Researcher free?

Yes. It is MIT licensed. You pay only for the LLM and search APIs you connect.

How do I run GPT Researcher?

As a Python package, a FastAPI server, or a Docker image. It can also run as an MCP server.

Which models does it support?

Any LLM, including OpenAI, Anthropic, Groq, Google, Mistral, DeepSeek and local models via Ollama.

More AI Research Agents (deep research)

See the full AI Research Agents (deep research) ranking