Ollama (on your Mac)
With Ollama, the model runs on your own Mac: nothing you send leaves it, and it works offline.
-
Download Ollama from ollama.com/download and open it.
-
In Terminal, download a model, for example:
Terminal window ollama pull gpt-oss:20bSmaller models such as
llama3.2suit Macs with less memory. -
In Ardra, choose Add a new model, then the Ollama on Mac shortcut. It fills in this address:
http://localhost:11434/v1 -
API Key: Ollama doesn’t need one, but the field can’t be empty, so type any word, such as
ollama. -
Click Find models and pick the model you downloaded.
-
Click Test, then Add model.