While experimenting with local LLMs on my Mac mini, I started wondering, "Can I use the Mac's built-in AI from my own tools as well?" I have run models like Qwen and Gemma before, but this time it is ...