running a local model via ollama

Before any Python script can talk to a model, the model itself has to be running and listening for requests. This is the part that happens entirely outside your code — in the terminal, one time (or on