AnotiaWang/deep-research-web-ui
(Supports DeepSeek R1) An AI-powered research assistant that performs iterative, deep research on any topic by combining search engines, web scraping, and large language models.
What it solves
This project provides a user-friendly web interface for the deep-research tool, allowing users to perform complex AI-driven research tasks without needing to interact with a command-line interface. It simplifies the deployment and configuration of deep research agents that can search the web and synthesize reports.
How it works
The UI acts as a frontend for the research process, supporting two deployment modes: Client Mode, where API keys are stored locally in the browser, and Server Mode, where keys are managed as environment variables on the server. It uses plain prompts to ensure compatibility with a wider range of AI providers (avoiding reliance on specific features like Structured Outputs) and integrates with various web search providers to gather information.
Who it’s for
Researchers, developers, and AI enthusiasts who want a visual way to manage deep research agents and export their findings into professional reports.
Highlights
- Search Visualization: Displays the research process as a tree structure to show how the agent explores topics.
- Broad Provider Support: Compatible with OpenAI-compatible APIs, DeepSeek, Ollama, OpenRouter, and more.
- Search Integration: Supports Tavily, Firecrawl, fastCRW, and Google PSE for web data retrieval.
- Real-time Feedback: Streams AI responses and updates the UI in real-time.
- Export Options: Allows exporting final research reports as Markdown or PDF files.
- Easy Deployment: Supports one-line Docker deployment and static site generation.
Related
- Project
- Project
- Project
- Project
- Project