The open-source workflow automation platform that connects your apps, builds AI agents, and runs on your own infrastructure. Everything you need to know, in one place.
n8n is an open-source workflow automation platform. It gives you a visual, node-based canvas where you connect apps, APIs, and services together into automated workflows. You can add logic, conditions, data transformations, custom code, and AI into every workflow you build. When something happens (a webhook fires, a schedule triggers, a form is submitted), your workflow runs automatically.
Think of it as the connective tissue between every tool your business or personal stack relies on. Instead of manually copying data from one system to another, writing one-off scripts, or paying someone to do repetitive digital tasks, you build it once on the n8n canvas and let it run. The platform handles over 500 built-in integrations out of the box, and anything with an API can be connected using the HTTP Request node.
n8n was created by Jan Oberhauser in Berlin and first released in 2019. Since then it has grown into one of the most popular open-source automation tools in the world, with over 50,000 GitHub stars and a $2.5 billion valuation after its Series C funding. The project is backed by Sequoia Capital, Felicis Ventures, and other major investors.
The name: n8n stands for "nodemation" (node + automation). The "8" represents the eight characters between the first "n" and the last "n". It is pronounced "n-eight-n."
The n8n editor is a visual canvas where you build workflows by placing nodes and connecting them with wires. Each node is a discrete operation: a trigger that starts the workflow, an action that does something (send an email, query a database, call an API), or a logic gate that routes data based on conditions.
Data flows through the workflow as items. Each item is a JSON object. When a trigger fires, it produces one or more items. Those items move through the connected nodes in sequence. Each node processes the incoming items, transforms them, filters them, or uses them to perform actions, then passes the results downstream.
There are two categories of nodes you will work with constantly:
Trigger nodes start your workflow. A Webhook Trigger listens for incoming HTTP requests. A Schedule Trigger fires at set intervals (every 5 minutes, every Monday at 9am). A Form Trigger presents a web form and fires when someone submits it. There are also app-specific triggers: "When a new row is added to Google Sheets," "When a message is posted in Slack," and dozens more.
Action nodes do the work. These include app integrations (Gmail, Slack, Notion, Airtable, HubSpot), data manipulation (Merge, Split, Set, Aggregate), logic (If, Switch, Loop), code execution (JavaScript, Python), HTTP requests, and AI operations (LLM calls, agent reasoning, vector store queries).
Inside every node, you can use expressions to reference data dynamically. Expressions let you pull values from previous nodes, apply functions, format strings, do math, and build dynamic payloads. The syntax is straightforward: {{ $json.email }} grabs the email field from the current item, {{ $('Previous Node').item.json.name }} references a specific field from a specific node.
n8n distinguishes between manual executions (you click "Test Workflow" in the editor to run it once and see the output) and production executions (the workflow is active and runs automatically when triggered). This separation is critical for building and debugging safely before anything goes live.
The range of what people build with n8n is enormous. At a high level, it falls into two categories: business process automation and technical infrastructure automation. Most teams end up doing both.
Best for: operations teams, founders, agencies, and anyone who wants to eliminate repetitive manual work.
Best for: developers, DevOps engineers, and technical teams who need reliable automation with real code access.
More use cases people build constantly:
This is where n8n separates itself from tools like Zapier and Make. While those platforms have added basic AI features, n8n has gone deep. The platform now functions as a full-featured AI agent builder, letting you create autonomous agents that reason, use tools, maintain memory, and take real actions across your connected services.
The core of this is the Agent node. You give it an LLM (the reasoning brain), a set of tools (actions it can take), memory (conversation context), and an output parser (structured responses). The agent receives input, decides which tools to call and in what order, executes them, evaluates the results, and responds. It is not following a fixed script. It is reasoning through the task.
n8n supports all the major LLM providers: OpenAI (GPT-4o, GPT-4.1, o3), Anthropic (Claude Opus, Sonnet, Haiku), Google (Gemini), and Ollama for running open-source models locally. You can swap providers without changing your workflow logic, which matters when pricing and capability shift rapidly.
Beyond agents, n8n supports AI chains for simpler, deterministic flows: summarize this document, classify this email, extract structured data from this text. Chains are faster and cheaper than agents when you know exactly what you need. The key distinction: agents choose their own path; chains follow yours.
For retrieval-augmented generation (RAG), n8n integrates with vector stores like Pinecone, Qdrant, Supabase, and in-memory options. You can build knowledge bases from your documents, embed them into vector space, and have your AI agents query them for context-aware answers. This is how you build AI systems that know about your specific business, products, and data.
Go deeper on AI in n8n:
n8n serves a remarkably wide audience. The visual canvas makes it accessible to people who have never written code, while the deep technical capabilities keep experienced developers engaged. Here is how different users approach it.
The common thread is this: if you have tasks that involve moving data between systems, making decisions based on information, or performing the same sequence of steps repeatedly, n8n is built to handle it.
n8n's source code is publicly available on GitHub, where it has accumulated over 50,000 stars. You can inspect the code, contribute to the project, build custom nodes, and deploy the entire platform on your own infrastructure. This is a significant differentiator from closed-source competitors like Zapier and Make, where you have zero visibility into what is running your workflows.
The license is a Sustainable Use License (formerly described as "fair-code"). You can use n8n freely for internal automation, personal projects, and business operations. The restrictions apply when you try to compete with n8n itself or embed it into a product you sell as a service. For the vast majority of users, this distinction never matters. For the edge cases, read the details: n8n License and Regulations →
Self-hosting means you run n8n on your own server: a VPS, a Docker container on your local machine, a Kubernetes cluster, or any infrastructure you control. Your data never leaves your environment. Your workflow definitions, credentials, execution logs, and results all live on your hardware. For organizations handling sensitive data, regulated industries, or anyone who simply wants full control, this is not a nice-to-have. It is a requirement.
The Community Edition is the free, self-hosted version. It includes the full canvas, all core nodes, unlimited workflows, and unlimited executions. Some features like advanced user management, SSO, and certain collaboration tools are reserved for the Enterprise tier, but for individuals and small teams, the Community Edition is genuinely complete.
This is one of the first decisions new users face. Both options give you the same core n8n platform. The difference is in who manages the infrastructure, what you pay, and what you trade off.
Best for: people who want to build workflows, not manage servers. Most beginners should start here.
See Cloud Plans →Best for: developers, privacy-conscious organizations, and anyone with the technical ability to run a server.
Full Comparison →Our recommendation: If you are just discovering n8n, start with Cloud. The free tier gives you enough to learn the platform, build real workflows, and decide if n8n fits your needs. You can always migrate to self-hosted later. If you already know you need data sovereignty or unlimited executions, go straight to self-hosted.
n8n is more than just the automation platform. There is a substantial ecosystem of community resources, learning materials, and extension points that make it easier to build, learn, and get help.
The best way to learn n8n is to build something. Not to read about it, not to watch someone else do it, but to open the canvas and create a workflow that solves a real problem you have. The platform rewards hands-on time more than anything else.
Here is the path we recommend:
Step 1: Get n8n running. The fastest route is n8n Cloud. Sign up and your workspace is ready in seconds. Free tier, no credit card required. If you prefer self-hosting, follow our setup guide for Docker, npm, or VPS deployment.
Step 2: Build your first workflow. Our complete guide walks you through building a real, functional workflow from scratch. It covers triggers, nodes, data flow, expressions, and everything you need to understand how the platform works: Build Your First n8n Workflow →
Step 3: Take the Academy courses. The n8n Academy at learn.n8n.io offers structured, free courses. Start with QS101 (Quickstart) for a 3 to 5 hour fast-track, or begin the Foundations program with N8N101 for deeper coverage.
Step 4: Start small, build something real. Pick the most annoying manual task in your day. Automate it. Then pick the next one. n8n is a tool that compounds over time. The more workflows you build, the more you see where automation fits, and the faster you move.
n8n is an open-source workflow automation platform with over 500 integrations, a visual node-based editor, and deep AI agent capabilities. It is self-hostable, free to use via the Community Edition, and backed by a $2.5B valuation. You can build everything from simple app-to-app connections to fully autonomous AI agents that reason, use tools, and take actions across your stack. Start with n8n Cloud for the fastest setup, or self-host for full control. Build your first workflow, take the free Academy courses, and go from there.