DeerFlow is an open-source research framework combining language models with tools like web search, crawling, and Python execution. It integrates with multiple MCP services, supports human-in-the-loop collaboration, and has both console and web UI options. Installation requires Python 3.12+, Node.js, and tools such as uv and pnpm for dependency management. The framework features modular multi-agent architecture, search engine compatibility, and text-to-speech capabilities for automated report synthesis into podcasts.