MindSearch
StaleDescription
An LLM-based multi-agent framework for web search engines, similar to Perplexity.ai Pro and SearchGPT, enabling intelligent web search.
Key Features
- Multi-agent web search framework mimicking human search patterns for deep, comprehensive results
- Supports multiple search engines: DuckDuckGo, Bing, Brave, Google, and Tencent
- Multiple frontend interfaces: React, Gradio, and Streamlit
- Asynchronous agent deployment for concurrent multi-query search
- Compatible with multiple LLM backends including InternLM2.5 and GPT-4
- FastAPI backend with WebSocket streaming support
Use Cases
💡 Deep research tasks requiring comprehensive web information gathering and synthesis
💡 Building Perplexity-style AI search engines with multi-source web search capabilities
💡 Academic research requiring extensive literature review and information aggregation
💡 Enterprise knowledge retrieval systems that need real-time web data integration
Strengths & Limitations
✅ Strengths
- • High community interest (6.9k stars)
- • Permissive open-source license (Apache-2.0)
- • Established track record (2 years in production)
- • Responsive to issues, low backlog
⚠️ Limitations
- • No updates in over 13 months
Categories
Quick Start
1. Clone the repo and install dependencies: `pip install -r requirements.txt`
2. Configure environment: copy `.env.example` to `.env` and add your API keys
3. Start the API server: `python -m mindsearch.app --lang en --model_format gpt4 --search_engine DuckDuckGoSearch --asy`
4. Launch a frontend (React, Gradio, or Streamlit) and enter your search query