Visual builder for LangChain style flows, agents and RAG pipelines
LangFlow is an open-source visual editor for LLM applications, originally built as a way to assemble LangChain components without writing the wiring by hand. You drag model, prompt, retriever, tool and memory components onto a canvas, connect them, and immediately test the result in a chat pane. Flows can be exported as JSON or as Python code, which means a prototype built visually can later be moved into a codebase. It self-hosts easily and supports local models, making it a common first stop for teams exploring retrieval and agent patterns.
Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.
prototyping rag apps, learning llm patterns, internal assistants and demos
If you're comparing similar products, check the alternatives below, or browse all tools in the AI Agents & Automation category.
Not to start. It helps when a flow grows, because the components mirror LangChain concepts and you will want to understand what each one is doing rather than only how it connects.
Yes. Flows export as Python code as well as JSON, which is the usual path from a visual prototype to something you maintain and deploy yourself.
It can serve production traffic, but scaling, secrets management and monitoring are your responsibility when self-hosting. Most teams use it to design and validate, then deploy in code.
Automation platform connecting thousands of apps, now with AI agents
Source-available workflow automation with native AI agent nodes
Open-source automation platform with a clean editor and AI pieces
Microsoft's framework for multi-agent conversations and tool use