# LangFlow
LangFlow is an open-source, Python-based, low-code framework for building AI applications, including [[AI Agents]], [[Retrieval-Augmented Generation (RAG)|RAG]] systems, chatbots, and agentic workflows.
It provides a visual drag-and-drop editor that simplifies the prototyping of AI workflows, enabling developers to quickly turn ideas into functional solutions. The visual approach makes complex workflows accessible to both technical and non-technical users.
## Key Features
- **Visual builder interface** for rapid prototyping and iteration
- **Source code access** to customize any component using [[Python]]
- **Interactive playground** to test and refine flows in real-time
- **Multi-agent orchestration** with conversation management
- **[[MCP]] support** (both as server and client)
- **Flexible deployment**: API, JSON export, or MCP server
- **Observability** through LangSmith, LangFuse, and other integrations
- **Extensibility** with custom Python components
## Architecture
LangFlow uses a flow-based architecture where you connect and configure component nodes. Each component represents a single step in the workflow. The platform supports:
- All major [[Large Language Models (LLMs)]]
- Popular [[Vector Databases]]
- A growing library of AI tools and integrations (Notion, GitHub, Slack, and many more)
## Use Cases
- Chatbots and conversational AI
- Document analysis systems
- Content generators
- RAG applications
- Agentic workflows
## Getting Started
LangFlow can be installed via pip or run via Docker:
```bash
# Install with uv (recommended)
uv pip install langflow -U
# Run
uv run langflow run
```
LangFlow also offers a Desktop application for Windows and macOS that bundles all dependencies.
License: MIT
## References
- Official Website: https://www.langflow.org/
- Documentation: https://docs.langflow.org/
- Source code: https://github.com/langflow-ai/langflow
- Discord: https://discord.gg/EqksyE2EX9