In this post I’m going to walk you through setting up N8N and Ollama on your fancy Apple Silicon Mac. Trust me, it’s not as scary as it sounds, and you’ll be running AI locally with GPU acceleration before you know it!
The cool part first… What makes this setup awesome is that you’ll be able to run everything locally on your Mac, using your GPU for some serious processing power. In my tests on an M3 MacBook Pro, this setup runs about twice as fast as the CPU-only version.
Let’s Break It Down
Step 1: Getting Docker Up and Running First things first, we need Docker. Just download it, install it, and create an account. It’s pretty straightforward – just follow the installer’s “next, next, next” prompts. Nothing tricky here!
Step 2: Enter the N8N Universe Here’s where it gets interesting. N8N has this fantastic self-hosted AI starter kit on GitHub that makes our lives so much easier. It’s basically a pre-packaged solution with most of what we need. Just clone the repository (fancy way of saying “download”), and you’re good to go.
Step 3: The Docker Container Magic Now, here’s a pro tip: we’re going to install the version without Ollama baked in. Why? Because we want to use your Mac’s GPU power, and the built-in version would only use your CPU. Trust me, this extra step is worth it for the performance boost!
Just run the Docker compose command in your terminal while in the starter kit folder, and let it do its thing. It’ll take about 5 minutes to pull everything it needs from the web. Go grab a coffee – you’ve earned it!
Step 4: The Ollama Integration This is the secret sauce that makes everything run smoothly with your GPU. Install Ollama separately (it’s super easy), and then we’ll connect it to N8N. The magic command you need to remember is pointing to https://host.docker.internal:11434 – this is what ties everything together.
Testing It All Out
Once everything’s set up, head to https://localhost:5678 in your browser. Create your login credentials, and you’re ready to rock! The best part? When you start chatting with the AI, everything’s running right on your Mac – no internet required for the actual processing.
Quick Troubleshooting Tips:
- Double-check your Docker container is actually running
- Remember to keep that terminal window open while using N8N
Why This Matters Running AI locally on your Mac with GPU support isn’t just cool – it’s practical. You get faster processing, better privacy (since everything’s local), and you’re making the most of that Apple Silicon chip you paid for!
Final Thoughts This setup might seem a bit involved at first, but it’s totally worth it. Once you get everything running, you’ll have a powerful AI setup right on your Mac, running at peak performance with GPU acceleration.
Have you tried setting this up on your Mac? Drop a comment below with your experience or any questions – I’d love to help out!
No Responses