How to Run Deep Research Locally: Docker Setup Tutorial

Getting Started with Deep Research

Deep Research is a TypeScript-based AI agent that autonomously conducts deep, iterative research. In this tutorial, you'll learn how to set it up locally using Docker, configure your preferred LLM, and run your first research session.

Prerequisites

Step 1: Run with Docker

The easiest way to run Deep Research is using the community Docker image:

docker run -d \
  --name deep-research \
  -p 8080:8080 \
  -v ./data/deep-research:/data \
  -e OPENAI_API_KEY=your_key_here \
  xiangfa/deep-research:latest

Step 2: Configure Your LLM

Deep Research supports multiple LLM providers. Set the environment variable for your preferred model:

• OpenAI (GPT-4): OPENAI_API_KEY
• Anthropic (Claude): ANTHROPIC_API_KEY
• OpenRouter (multi-model): OPENROUTER_API_KEY

Step 3: Run Your First Research

Once running, submit a research topic through the web interface or API. The agent will:

1. Generate initial search queries — broad exploration of your topic
2. Scrape and analyze results — extract key information from each source
3. Identify knowledge gaps — refine queries to dig deeper into uncovered areas
4. Repeat iteratively — each pass adds new depth and context
5. Produce a final report — a structured, citation-rich research document

Real-World Use Cases

Deep Research excels at tasks that require multi-source synthesis:

• Market research: Analyze competitors and market trends
• Academic literature reviews: Explore papers, blogs, and technical docs
• Technical investigations: Compare frameworks, tools, and architectures
• Due diligence: Deep-dive into companies, technologies, or regulations