AI Agent Platforms / All platforms / Langflow
Langflow
Visual and code-based platform for building agents and workflows served as APIs or MCP servers.
Langflow is a platform for building AI agents and workflows through a visual builder with full Python source access to components. Developers use it to prototype flows in an interactive playground and deploy them as APIs or MCP servers.
| Category | Agent builders and frameworks |
| License | MIT |
| Hosting | self-hosted; vendor-hosted available |
| Repository | github.com/langflow-ai/langflow |
| Website | docs.langflow.org |
| Last verified | 2026-09-02 |
Documented features
- Builds flows in a visual builder interface
- Allows customizing any component in Python through source code access
- Tests flows in an interactive playground with step-by-step control
- Deploys flows as an API or exports them as JSON for Python apps
- Deploys flows as an MCP server so they become tools for MCP clients
- Orchestrates multiple agents with conversation management and retrieval
- Supports major LLMs and vector databases
- Integrates observability tools including LangSmith and LangFuse
- Ships a desktop application for Windows and macOS
Feature summaries come from the project's own repository and documentation and describe the software as documented on the verification date, not an evaluation.