Ollama has become the default answer when someone asks how to run a local LLM, and for good reason. It's easy, it works across platforms, and it hides enough of the ugly parts that you can go from ...