GPT Researcher

What is GPT Researcher?

GPT Researcher is an open source autonomous agent for online research. You give it a question and it splits the work into sub-questions, searches for each one in parallel, reads more than 20 sources, filters out passages that do not answer the question, and writes a report with citations. It can also research your own PDFs, Word files, spreadsheets and Markdown, alone or combined with web results, and a deep research mode branches into follow-up questions. Reports export to PDF, Word or Markdown. It is available as a Python package, a FastAPI server you host yourself, a Docker image, or via the gptr-mcp server, with the LLM and search provider of your choice. Analysts, students and developers use it for market research, literature reviews and due-diligence briefs. The software is free, so your only cost is the LLM and search API usage you configure.

Use cases

• Write cited research reports
• Run market and competitor research
• Compile literature reviews
• Prepare due-diligence briefs
• Research across your own documents
• Give other agents a deep research tool
• Export reports to PDF or Word

Who it's for

• Analysts
• Researchers
• Students
• Developers
• Product teams

Why use GPT Researcher

• Free open source code with no subscription, so spend is limited to the model and search calls you choose.
• Swappable LLMs and retrievers let you pick cost and quality, or run fully on-premises with a local model and SearXNG.
• Every claim in a report carries a citation, which makes checking the sources faster.
• Can be embedded in other agents and IDEs through its MCP server, API and Python package.
• Handles web sources and private documents in the same research run.

GPT Researcher features

• Planner and execution agents that run research in parallel
• Reads 20+ web sources per report with citations
• Deep research mode that branches into sub-questions
• Research over local PDF, Word, spreadsheet and Markdown files
• Quick search and long-form report writing tools
• Works with OpenAI, Anthropic, Google, Mistral, DeepSeek and local models
• Retrievers including Tavily, Bing, Google, DuckDuckGo and SearXNG
• MCP server (gptr-mcp) and MCP data sources
• Streaming over WebSocket and server-sent events
• Report export to PDF, Word and Markdown
• Docker image and FastAPI server for self-hosting

GPT Researcher integrations

• OpenAI
• Anthropic
• Google
• Mistral
• DeepSeek
• Ollama
• Tavily
• Bing
• DuckDuckGo
• SearXNG
• LangGraph
• Claude Desktop
• Cursor

How to get started with GPT Researcher

1. Install the package with pip, or start the Docker image or FastAPI server.
2. Add an LLM API key and a search API key such as Tavily, or point it at local documents.
3. Enter a query and pick a report type; the agent breaks it into focused sub-questions.
4. Crawler agents search and summarize sources for each sub-question while tracking where each fact came from.
5. The planner filters and combines the findings into a cited report that you can export.

GPT Researcher limitations

• Requires Python 3.11 or newer plus at least one LLM API key and one search API key
• A deep research run typically takes 30 to 60 seconds per query
• Output quality depends on the LLM and retriever you configure
• Users are responsible for fact-checking the cited reports
• There is no first-party hosted service; gptr.dev is a landing page

GPT Researcher pricing

GPT Researcher is open source, so you can use and self-host it for free.

GPT Researcher plans

• Open Source

  $0

  • Python package and gptr-mcp server
  • Self-host on your machine or container platform
  • Any supported LLM provider and retriever
  • PDF, DOCX and Markdown report export
  • Multi-agent orchestration

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

GPT Researcher alternatives

The main GPT Researcher alternatives listed on Debuly are Perplexity, Elicit, Local Deep Research, Chart Library, OpenResearch and VERDICT. The table compares pricing and platforms; order is A–Z.

1. 

   Chart Library

   Lookup tool and MCP/API service that finds historical analogs for a stock's recent chart and shows what followed.

   Research and searchDeveloper platformsAutomation and agents
   FreeVisit Chart Library (official site)
2. 

   Elicit

   AI research assistant that searches 138 million papers, extracts data into tables and runs systematic reviews and cited reports.

   Research and searchProductivityWriting
   FreemiumVisit Elicit (official site)
3. LD

   Local Deep Research

   Self-hosted AI research assistant that runs agentic deep research with local or cloud LLMs, many search engines and cited reports, with encrypted storage.

   Research and searchDeveloper platforms
   Open sourceVisit Local Deep Research (official site)
4. 

   OpenResearch

   Local-first workspace from alphaXiv that turns coding agents into research agents and rents GPU compute for experiments

   Research and searchAutomation and agentsCoding
   Open sourceVisit OpenResearch (official site)
5. 

   Perplexity

   AI answer engine that searches the web, cites sources inline and offers deep research, multi-step tasks and a choice of top models on paid plans.

   Research and searchChat assistantsProductivity
   FreemiumVisit Perplexity (official site)
6. VE

   VERDICT

   VERDICT is an AI legal research platform that searches court cases, drafts documents from templates and helps you find an attorney.

   Research and searchWriting
   FreemiumVisit VERDICT (official site)

GPT Researcher FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from gptr.dev · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.