Langflow
An open-source AI workflow builder with local and Docker deployment, Windows and macOS desktop apps, Ollama support, and API and MCP servers.
Langflow is a visual builder for developers creating AI agents and retrieval-augmented generation (RAG) applications. You can run it locally or on your own server, with Docker support and desktop apps for Windows and macOS. The open-source software uses the MIT license. A hosted cloud offering provides a separate deployment option.
Its drag-and-drop editor connects models, data sources and tools into workflows. Prebuilt components and reusable flows give you a starting point, while Python customization lets you change components when the visual editor doesn't cover your needs. An interactive playground lets you test flows and inspect their behavior step by step.
Model choice stays flexible. Langflow connects to Ollama for local LLM use and to cloud providers including OpenAI, Anthropic and Amazon Bedrock. Retrieval workflows can use vector stores such as Qdrant, Milvus and Weaviate, alongside data integrations including Google Drive and Confluence. Running the builder locally is separate from choosing a local or cloud model backend.
You can build a single agent or coordinate multiple agents with conversation management and retrieval. Agents can use workflow components as tools. Built-in API and MCP servers expose flows to other applications and MCP clients, so the same visual workflow can become part of an existing app. Flows also export as JSON for Python applications, and LangSmith and LangFuse integrations support observability.
