deep-research

Overview

Deep Research is a powerful MCP server that enables in-depth research reports using various AI models. It combines advanced thinking and task models with internet connectivity to provide comprehensive analysis on diverse topics. The server supports multiple LLM providers including Gemini, OpenAI, Anthropic, and others, along with various search engines like Searxng, Tavily, and Firecrawl. All processing can be done locally to ensure privacy while still delivering rapid research results in about 2 minutes.

Try asking AI

After installing, here are 7 things you can ask your AI assistant:

• deep_research

  Perform deep research on a topic using various AI models and search providers

  Try:Research the latest trends in artificial intelligenceTry:Conduct an analysis of sustainable energy solutionsTry:Research the impact of remote work on productivity

Note: The README mentions MCP server support but doesn't provide explicit tool documentation. The tool name 'deep_research' was inferred from the project description and API documentation indicating it's the core functionality.

Comparable tools

Installation

docker pull xiangfa/deep-research:latest
docker run -d --name deep-research -p 3333:3000 xiangfa/deep-research

git clone https://github.com/u14app/deep-research.git
cd deep-research
pnpm install
# Copy env.tpl to .env and configure your API keys
pnpm dev

{
  "mcpServers": {
    "deep-research": {
      "command": "node",
      "args": ["path/to/deep-research/server.js"]
    }
  }
}

FAQ

What LLM providers are supported?

The server supports Gemini, OpenAI, Anthropic, Deepseek, Grok, Mistral, Azure OpenAI, any OpenAI Compatible LLMs, OpenRouter, Ollama, and others.

How is privacy ensured?

All data is processed and stored locally by default, with options for server-side API calls when configured properly.

Can I use this as an MCP server?

Yes, the project specifically supports MCP server functionality alongside SSE API for real-time research tasks.

Compare deep-research with