Langflow
Langflow is an open-source platform under MIT for building and deploying AI-powered agents and workflows visually. It supports all major language models and vector databases, lets you edit any component in Python, runs multi-agent orchestration, and turns every flow into an API or an MCP server that other tools can call. It installs with pip or Docker, or as a desktop app for Windows and macOS.
What does Langflow do?
Langflow is "a powerful platform for building and deploying AI-powered agents and workflows", with "a visual authoring experience and built-in API and MCP servers that turn every workflow into a tool." It "supports all major LLMs, vector databases and a growing library of AI tools", and "source code access lets you customize any component using Python."
An interactive playground runs flows step by step, and multi-agent orchestration handles conversation management and retrieval. Finished flows deploy as an API, export as JSON for Python apps, or "deploy as an MCP server and turn your flows into tools for MCP clients." Observability integrates with LangSmith, Langfuse and others. The README says it is completely open source and deployable to all major clouds.
Key facts
- Licence: MIT, copyright Langflow, with no other licence files at the top of the repository.
- Visual builder for AI agents and workflows, with an interactive step-by-step playground.
- Supports all major LLMs and vector databases, by the README; providers are not listed by name.
- Every component can be customised in Python.
- Multi-agent orchestration with conversation management and retrieval.
- Deploy each flow as an API, export as JSON, or serve it as an MCP server.
- Install: uv pip install langflow then uv run langflow run on port 7860, with Python 3.10 to 3.14; a Docker image; a desktop app for Windows and macOS.
- Latest tagged release when read: v1.12.2 on 2026-09-15; 154,959 stars on 2026-09-17.
What does it replace, and where does it fall short?
Langflow replaces Zapier and Make where the workflow is an AI agent or a chain of model calls, under a plain MIT licence, and where the result should become an API or a tool for other agents. It is the MIT pick for AI workflows on the open-source alternatives to Make page.
Where it falls short: it is built around AI components rather than the thousands of app-to-app connectors Zapier and Make sell, and the README gives no count of integrations. Triggers and webhooks are not described in the README. For broad app integrations, n8n or Activepieces.
How does Langflow run?
uv pip install langflow, then uv run langflow run, which serves on port 7860; it needs Python 3.10 to 3.14. Or docker run the langflowai/langflow image. Langflow Desktop is available for Windows and macOS. The README describes no paid tier and says it deploys to all major clouds.
Who is Langflow for?
A developer or team building AI agents and model workflows who wants to expose them as APIs or MCP tools under MIT. Someone automating routine app-to-app tasks should look at Activepieces or n8n.
What limits does the README state?
None stated in the README.
Questions people ask
Is Langflow open source?
Yes. Langflow is MIT licensed, and the README says it is completely open source and deployable to all major clouds. There are no enterprise folders or added conditions in the repository.
Can Langflow turn a workflow into an MCP server?
Yes. The README says you can deploy a flow as an MCP server and turn your flows into tools for MCP clients, alongside deploying it as an API or exporting it as JSON for Python apps.
Which models does Langflow support?
All major LLMs and vector databases, by its README, which does not list them by name. Components are editable in Python, so any model with a Python client can be wired in.
How does Langflow compare with Make?
Langflow is a visual builder for AI agents and model workflows under MIT, with every flow deployable as an API or MCP server. Make is a hosted platform for connecting apps and AI models visually. The open-source alternatives to Make page compares Langflow with n8n, Dify, Activepieces, Windmill and Huginn.
Is there a desktop version of Langflow?
Yes. Langflow Desktop is available for Windows and macOS, by the README. Otherwise it runs anywhere Python 3.10 to 3.14 or Docker runs, served in a browser on port 7860.
Sources
- Langflow README and MIT LICENSE: github.com/langflow-ai/langflow, read 2026-09-17.
Compared with the others
On the open-source alternatives to Zapier page, Langflow is the pick for visual ai flows as apis or mcp tools. MIT visual builder for agents and workflows, with Python-editable components and every flow deployable as an API or MCP server.
Also on that page: Activepieces for a no-code builder under mit, n8n for the largest integration library, Windmill for code-first automation for developers.
On the open-source alternatives to Make page, Langflow is the pick for visual ai flows as apis or mcp tools. MIT visual builder for agents and workflows, with Python-editable components and every flow deployable as an API or MCP server.
Also on that page: n8n for the largest integration library, Activepieces for a no-code builder under mit, Dify for ai-first workflows and agents.
More workflow automation
A fair-code workflow automation platform with 1,500+ integrations, 9,000+ templates, AI agents over OpenAI, Anthropic, Google or open models, and code steps in JavaScript and Python, under the Sustainable Use License.
An LLM app development platform with visual AI workflows, RAG pipelines, sandboxed agents, MCP tools and hundreds of models, under Apache-2.0 plus conditions that forbid multi-tenant use and removing the logo without permission.
An MIT system, running since 2013, for building agents that watch the web and act for you, which its README calls a hackable version of IFTTT or Zapier on your own server, with webhooks, scraping and JavaScript steps.
An MIT automation builder its README calls an open-source replacement for Zapier, with hundreds of TypeScript pieces that double as MCP servers, AI agents, human approval steps and a no-code builder.
Added September 17, 2026. Every claim above comes from the project's README, LICENSE or model card, read on September 17, 2026, or from the GitHub API on the date shown in the panel. Found an error? Write to reveneau@licheo.com and it is fixed in the next weekly pass. Repository: github.com/langflow-ai/langflow.
Running one of these inside your own environment, with your own data and your own security rules, is the kind of work Reveneau does. Read how a forward deployed engagement works.