This project provides a web UI for an AI-powered research assistant, enabling iterative, deep research on any topic by integrating search engines, web scraping, and large language models. It targets users who need to conduct thorough research and generate reports, offering a real-time, visual, and exportable workflow.

How It Works

The system orchestrates a multi-stage research process, starting with user-defined topics and progressively refining queries through LLM interactions. It visualizes this process as a tree, showing search queries, scraped content, and AI-generated insights. The architecture prioritizes local execution within the browser for security and real-time feedback, supporting a wide array of AI and web search providers through flexible prompt engineering.

Quick Start & Requirements

• Install/Run: docker run -p 3000:3000 --name deep-research-web -d anotia/deep-research-web:latest
• Prerequisites: Docker. For development: Node.js, pnpm.
• Links: Live Demo

Highlighted Details

• Supports multiple AI providers (OpenAI compatible, DeepSeek, Ollama, etc.) and web search providers (Tavily, Firecrawl, Google PSE).
• Real-time streaming of AI responses and UI updates.
• Research process visualized as a tree structure with multi-language search support.
• Exportable research reports in Markdown/PDF format.

Maintenance & Community

The project is actively maintained with frequent updates addressing provider support, UI improvements, and bug fixes. Links to community channels are not explicitly provided in the README.

Licensing & Compatibility

• License: MIT
• Compatibility: Permissive MIT license allows for commercial use and integration with closed-source projects.

Limitations & Caveats

While supporting many providers, the effectiveness of the research is dependent on the quality of the underlying LLMs and search services used. Some UI fixes and improvements are recent, suggesting potential for minor regressions.