How to install n8n on macOS?

The most reliable way to install n8n on macOS today is to run it with Docker. It isolates all dependencies, avoids Node.js version conflicts, and behaves very close to how n8n runs in real production. With Docker installed, you can start n8n with a single command and access it in your browser.

 

Recommended: Install n8n on macOS using Docker

 

Below is the simplest working setup that runs n8n locally, stores your workflows on your Mac (so they don’t disappear when you restart), and behaves predictably:

• Install Docker Desktop for macOS from the official site
• Create a folder to store n8n’s data (SQLite DB and files)
• Run an n8n Docker container that mounts that folder
• Open n8n in your browser

This command starts n8n on port 5678 and keeps all your workflows persistent on your Mac:

docker run -it \
  --name n8n \
  -p 5678:5678 \
  -v ~/n8n_data:/home/node/.n8n \
  n8nio/n8n

Once the container finishes starting, open:

http://localhost:5678

This gives you a full n8n instance running exactly like a real self‑hosted installation.

 

Alternative: Install n8n on macOS using Node.js

 

You can install n8n directly with Node.js, but it’s not what I recommend for long-term use. Still, for simple experimentation or if you are allergic to Docker, here is the valid method. First, install a supported Node.js LTS version (n8n typically supports the active LTS). Then install n8n globally:

npm install -g n8n

Start n8n:

n8n

Again, you access it at:

http://localhost:5678

Just remember: on macOS this setup stores n8n’s internal SQLite DB inside your global npm folder. It works, but it’s easy to break if you upgrade Node or reinstall global packages. Docker avoids those problems.

 

What actually happens under the hood

 

Regardless of whether you run Docker or Node directly, n8n boots an internal server (Express under the hood) on port 5678. That server hosts both the UI and REST API. n8n stores your workflows and credentials in a small SQLite database unless you configure something like Postgres. The mounted folder in the Docker example above holds that data, so you don’t lose work on restart.

If you later want production‑grade behavior (auto-restart, long-running environments, Postgres, Redis, queues), Docker is the path you’re already on. That’s why starting with Docker on macOS is not just easiest — it builds habits aligned with how real n8n deployments work.