About Langflow
Langflow is an open source visual builder for AI agents and workflows. You arrange components on a canvas, connect them, and run the result as a flow that can also be served as an API.
What it does
Each block in Langflow stands for a model, a data source, a tool or a logic step, and you connect the blocks with lines. You can test a flow in the built-in playground and swap models to compare their output. Because the underlying code is Python, you can write custom components when the visual blocks are not enough.
Key features
- Drag-and-drop component canvas
- Swap models and compare outputs
- Flows served as an API
- Python code for custom components
- Library of tools and data connectors
Who it's for
Developers and non-programmers who want to prototype agent or retrieval setups quickly. Teams that want to run the same flow on their own machines or in a hosted cloud can use the project both ways.
Pricing and good to know
Langflow is released under the MIT license and can be self-hosted at no software cost. A free cloud account is also offered for hosted deployments. Hosted terms can change, so check them before you depend on one.
Highlights
- Drag-and-drop component canvas
- Swap models and compare outputs
- Flows served as an API
- Python code for custom components
- Library of tools and data connectors
