Skip to main content

Set Up Ollama Securely at Home: Step by Step

This walkthrough installs Ollama (a tool that runs AI models on your own computer) and confirms it is only reachable from your own device, the safe default. By the end you will have a working local AI assistant that never sends your data anywhere.

Reading is great. Tracking makes it stick. Sign up for a free Dashboard to tick off tasks and see your Security Score.

Get my free Dashboard →
CE: Firewalls⏱ 15 minutes
A laptop screen showing lines of code

Cyber Nova AI helps you prepare for Cyber Essentials. It doesn't award or certify it. Follow the steps here to get ready, then apply for certification through an accredited Cyber Essentials body when you're set.

What you will need

  • A computer running Windows, Mac, or Linux with at least 8GB of RAM
  • About 15 minutes
  • Basic comfort opening a terminal or command prompt
1

Download and install Ollama

Go to ollama.com and download the installer for your operating system. Run it like any other program.

2

Run your first model

Open a terminal and run:

ollama run llama3.2

Note: This downloads the model the first time, which can take a few minutes.

3

Confirm Ollama is only reachable from your own computer

By default it only listens on your own device, not your home network or the internet. Check with:

echo $OLLAMA_HOST

Note: On Windows Command Prompt, use "echo %OLLAMA_HOST%" instead. Nothing returned means it is using the safe default.

4

Only if you need it: allow access from other devices on your home network

Set OLLAMA_HOST to make it reachable from your own network:

OLLAMA_HOST=0.0.0.0:11434 ollama serve

Note: This makes Ollama reachable from your whole home network. It still has no password, so only do this on a network you trust completely, and never forward this port through your router to the internet.

5

If you ever need remote access, add a password in front of it

Ollama has no login of its own. Put a reverse proxy in front of it with a username and password, rather than exposing it directly.

Note: A more advanced step, covered in a separate guide. Most home use stops at step 3.

What you will have at the end

A working local AI assistant, running entirely on your own hardware, reachable only from your own computer by default.

Why the default setting matters

Ollama has no username, password, or login screen. Anything that can reach it can use your AI models, or delete them, with no credentials at all. In January 2026, security researchers found around 175,000 Ollama installations exposed to the open internet, almost all because someone changed the access setting without adding any protection in front of it. Leaving Ollama on its default setting avoids this entirely.

Frequently asked questions

Is Ollama safe to run at home?

Yes, by default it only accepts connections from your own computer.

Can I use Ollama from my phone on the same WiFi?

Yes, using the network access step above, but only on a home network you trust.

What models can I run?

Smaller models like Llama 3.2 run well on most modern computers; larger ones need more RAM and ideally a dedicated graphics card.

Track your AI safety progress. Free.

Create a free account to tick off tasks, see your Security Score improve, and know exactly what you've done and what's still to do.

Start your free security check
← Back to AI Homelab Projects Hub