langflow-ai/langflow

Langflow is an open-source, low-code platform for building and deploying AI-powered agents and workflows. It pairs a drag-and-drop visual builder with full Python source access, so every flow can be tested in an interactive playground, exposed as an API, or published as an MCP server for use in any stack. With batteries-included support for all major LLMs, vector databases, and a large library of AI tools, it moves AI prototypes from notebook to production quickly.

Key Highlights

• Visual builder with reusable components, plus Python source access to customize any node.
• Interactive playground for step-by-step testing, debugging, and iteration on flows.
• Multi-agent orchestration with conversation management, memory, and retrieval (RAG).
• Deploy anywhere: export as JSON, serve as a REST API, or expose flows as MCP server tools.
• Enterprise-ready observability (LangSmith, LangFuse), security, and scalability.

Best For

Developers, AI engineers, and product teams who want to prototype and ship agentic AI applications fast without giving up code-level control. It suits both Python-savvy builders and teams that prefer a visual, collaborative workflow.