Repolode
Submit
langflow-ai

langflow-ai/langflow

A visual platform for building, testing and deploying AI agents and workflows, with built-in API and MCP server exposure.

View on GitHubWebsite
Stars
155,863
Forks
10,262
Open issues
1,161
Last commit
6 hours ago

What is it?

Langflow is a platform for building AI-powered agents and workflows. It offers a visual builder for assembling flows from components, along with a Python source-code layer for customizing any component. Each finished workflow can be exposed as an API endpoint, exported as JSON for use in Python applications, or deployed as an MCP server so that its flows act as tools for MCP clients.

The interactive playground lets developers run and step through flows while refining them. The project supports multi-agent orchestration with conversation management and retrieval, and it lists integrations for major LLMs, vector databases and observability tools such as LangSmith and LangFuse. A desktop build is available for Windows and macOS, and the package can also be installed locally with uv, run from source or launched in Docker.

Key features

  • 01Visual drag-and-drop builder for agent and workflow design
  • 02Python source access for customizing components
  • 03Interactive playground for step-by-step testing
  • 04Deploy flows as REST APIs or as MCP servers
  • 05Integrations with major LLMs, vector databases and LangSmith/LangFuse

Use cases

  • Prototyping a RAG or chatbot pipeline without writing orchestration code from scratch
  • Exposing an agent workflow as a tool for MCP-compatible clients
  • Adding LLM-powered steps to an existing application through a generated API

Quick start

  1. 1

    Install Langflow with uv from a fresh directory

    uv pip install langflow -U
  2. 2

    Start the Langflow server

    uv run langflow run
  3. 3

    Open http://127.0.0.1:7860 in your browser (set LANGFLOW_SUPERUSER_PASSWORD first if authentication is enabled)

Things to know

  • Requires Python 3.10–3.14 and the uv package manager.
  • Official releases disable auto-login, so you must set an admin password before first use.

Topics

#ai-agents#llm#workflow#visual-programming#mcp#rag#python#agents#chatgpt#generative-ai

FAQ

What is langflow?
A visual platform for building, testing and deploying AI agents and workflows, with built-in API and MCP server exposure. Langflow is a platform for building AI-powered agents and workflows. It offers a visual builder for assembling flows from components, along with a Python source-code layer for customizing any component. Each finished workflow can be exposed as an API endpoint, exported as JSON for use in Python applications, or deployed as an MCP server so that its flows act as tools for MCP clients.
Is langflow free to use? What license does it have?
langflow is open source under the MIT license. Check the license terms before using it in commercial products.
Is langflow still maintained?
Actively maintained. The last commit to langflow was on Oct 20, 2026.
What can I use langflow for?
Prototyping a RAG or chatbot pipeline without writing orchestration code from scratch Exposing an agent workflow as a tool for MCP-compatible clients Adding LLM-powered steps to an existing application through a generated API
How do I install langflow?
Install Langflow with uv from a fresh directory: uv pip install langflow -U Start the Langflow server: uv run langflow run Open http://127.0.0.1:7860 in your browser (set LANGFLOW_SUPERUSER_PASSWORD first if authentication is enabled)
Where can I get langflow?
The source code and installation instructions are on GitHub: https://github.com/langflow-ai/langflow

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects