Use Ollama or LM Studio (private and free)

Want me to be clever and keep everything on your own computer, for free? Run a local AI model with Ollama or LM Studio, and point me at it. Nothing leaves your machine. It's my favourite option for the privacy-minded.

What it is

Ollama and LM Studio are free apps that run an AI model right on your computer. Once one is running, you tell me its address, and I use it as my brain — without sending your saves to anyone.

Why you'll love it

• Private. Your data never leaves your machine.
• Free. No per-use costs once it's set up.
• Clever. Modern local models are surprisingly capable.

How to use it

1. Install Ollama (ollama.com) or LM Studio (lmstudio.ai).
2. Download a model in that app (start with a small, well-known one).
3. Start the app so it's serving locally — usually at an address like http://localhost:11434 (Ollama) or http://localhost:1234 (LM Studio).
4. In Bommel Bot, open Settings → AI, choose the local/OpenAI-compatible option, and paste that address.
5. Pick your downloaded model from the list and save.

First time with local AI? Start with a smaller model. It'll be quick and light on your computer, and you can always try a bigger, brainier one later.

Good to know

• This is Free on Bommel Bot's side; you just need a computer that can run the model.
• If you're using the deployed website (over HTTPS) and your model runs on your own machine, there are a couple of small connection settings to get them talking — your Bommel Bot setup notes cover this.
• A local model needs to be running for me to use it. If I can't reach it, I'll quietly fall back to my built-in brain.

Where to next

• Choose the brain that helps me think
• Use OpenAI, Claude, Gemini, Grok, or Perplexity
• Sign-in, verification, and keeping you safe