Loading page
Loading page
AI
Build AI agents and RAG workflows on a visual canvas, then serve each one as an API or an MCP server.
Langflow is a platform for building AI agents and workflows. You drag components onto a canvas, such as a model, a prompt, a vector store or a tool, and connect them into a flow. A playground lets you run the flow and watch each step.
Every flow can become an API endpoint or an MCP server, so other apps and AI tools can call it. When the built-in components are not enough, you can open any component's Python code and change it.
Langflow suits developers and AI teams who want to prototype quickly and still ship something real. It also helps product people and analysts who understand the logic of an AI workflow but prefer not to write all the glue code.
AI flows often touch private documents and API keys. Running Langflow on your own service keeps both in your own database. You choose the model providers, and you can point flows at a model you host yourself.
The trade-offs are cost and care. You still pay your model providers for the calls your flows make. The template uses the latest image, so updates can change behavior. Pin a version once things work, back up PostgreSQL, and set a superuser password so the editor is not open to anyone with the URL.
Containers: langflowai/langflow:latest, postgres:16
Step 1
Choose the Starter plan. It costs $20 a month and runs up to 3 apps.
Step 2
Open App templates in the dashboard, choose Langflow, and click Deploy. We set up the containers, secrets and domain.
Step 3
Wait a few minutes for the first deploy, then open your new URL and finish the Langflow setup screen.