Guide · Open source

How to use Langflow: install, run and deploy AI workflows

Updated 2026-10-29 · written with AI from the project's own documentation and checked against our catalogue data

Langflow is an open-source platform for building and deploying AI-powered agents and workflows. It pairs a visual editor with built-in API and MCP servers, so each workflow can become a tool used by other applications. This guide shows you how to install it, start it safely, and understand what you can do once it is running.

Langflowlangflow-ai/langflow · ★ 156k · MIT Install options & downloads

Quick start

1Check your Python version. Confirm you have Python between 3.10 and 3.14. Install uv if you do not already have it.
2Install the package. Run uv pip install langflow -U from a fresh directory to get the latest Langflow package.
3Set the admin password. Set LANGFLOW_SUPERUSER_PASSWORD before starting, since official releases disable automatic login.
4Start the server. Run uv run langflow run. Langflow starts at http://127.0.0.1:7860.
5Sign in. Open the local address in your browser and sign in as langflow, or your configured LANGFLOW_SUPERUSER, with the password you set.
6Build and test a flow. Assemble components in the visual builder and test them in the playground before deploying.

What Langflow requires before you install

Langflow is a Python package. The README says it requires Python 3.10 to 3.14 and recommends uv as the package manager. Check your Python version before you start.

The README also mentions Langflow Desktop, an installer with all dependencies included, available for Windows and macOS. If you do not want to manage Python environments, that is the option the README describes as the easiest way to start.

Python 3.10 through 3.14
uv (recommended package manager)
Or Langflow Desktop on Windows or macOS

Install Langflow with uv

From a fresh directory, run the install command. The README says this installs the latest Langflow package.

For more detail, the README points to the Langflow documentation on installing and running the OSS Python package.

uv pip install langflow -U

Official release packages and tagged source archives disable automatic login by default. Set LANGFLOW_SUPERUSER_PASSWORD before starting Langflow. Then sign in with the username langflow, or with the value of LANGFLOW_SUPERUSER if you configured one.

Run the start command below. Langflow starts at http://127.0.0.1:7860 according to the README.

Set LANGFLOW_SUPERUSER_PASSWORD first, or sign-in will not work with the default setup.
Development branches keep automatic login on by default, so check which version you are running.

uv run langflow run

Run Langflow in Docker

The README includes a Docker option. Set and export a strong password before starting the container. The example uses a Bash prompt to read the password without displaying it.

Once the container runs, Langflow is available at http://localhost:7860/. Sign in as langflow with the password you entered. The README notes that official Docker images require authentication by default. For configuration options, see the Docker deployment guide linked in the README.

read -rsp "Langflow administrator password: " LANGFLOW_SUPERUSER_PASSWORD; echo
export LANGFLOW_SUPERUSER_PASSWORD
docker run -p 7860:7860 -e LANGFLOW_SUPERUSER_PASSWORD langflowai/langflow:latest

Build flows in the visual editor and test them

Langflow provides a visual builder for assembling flows, and an interactive playground for testing and refining them with step-by-step control. The README also says you can customize any component using Python source code.

When a flow works, you can deploy it as an API, export it as JSON for Python apps, or deploy it as an MCP server so MCP clients can use it as a tool.

Visual builder for quick iteration
Interactive playground with step-by-step control
Source code access to customize components in Python
Export flows as JSON or deploy them as an API or MCP server

Prepare Langflow for a public server

The README says Langflow is ready for deployment to major clouds and links to deployment guides. Before exposing an instance publicly, plan authentication. Auto-login is disabled in official releases, and a password must be set.

The README points to a Security Policy and contributing guidelines for further information.

Set a strong LANGFLOW_SUPERUSER_PASSWORD
Read the Langflow deployment guides before going live
Review the SECURITY.md file in the repository

What people use it for

Prototype an LLM workflow visually, then move it into Python code
Build a multi-agent or retrieval-based assistant
Expose a finished workflow as an API for other applications
Turn a workflow into an MCP server so MCP clients can call it as a tool

Frequently asked questions

Is Langflow free to use?

Langflow is released under the MIT licence, as shown in the repository. The README describes it as completely open source. Check the licence file in the repository for the full terms.

What Python version does Langflow need?

The README lists Python 3.10 through 3.14 as required. Using a version outside that range is not supported according to the README.

How do I install Langflow?

Run uv pip install langflow -U from a fresh directory, then start it with uv run langflow run. The README recommends uv as the package manager. Windows and macOS users can also use the Langflow Desktop installer.

Why can't I log in to Langflow?

Official releases disable automatic login. Set LANGFLOW_SUPERUSER_PASSWORD before starting Langflow, then sign in as langflow or your configured LANGFLOW_SUPERUSER.

What is Langflow used for?

The README describes it as a platform for building and deploying AI-powered agents and workflows. Each workflow can be exposed as an API or as an MCP server for other applications and tools.

How does Langflow compare to Flowise, Dify or n8n?

Langflow is a broader platform with many components to configure, so it may suit developers more than people who want a simple chat app. Flowise, Dify, n8n and LlamaIndex are comparable projects, and this page does not benchmark them against each other.

New agent skills, repos and AI tools — every week

The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.

Our verdict

Langflow suits developers and teams who want to prototype LLM workflows visually, then deploy them as APIs or MCP tools. It is a broad platform with many components, so people who only want a simple chat app with no setup should choose something lighter. Set a password before any public deployment.

Install options & downloads