6-2 Running an LLM on Your Own Machine The Ollama Way Hands-On

Last lecture we built the mental model — tokens, parameters, context windows, sampling. All theory. This lecture, we stop talking and we start running. By the end of the next fifteen minutes you will have a real large language model sitting on your laptop, responding to prompts you type, with no internet connection required. The tool that makes this absurdly easy is called Ollama, and once you see it work the first time, you will wonder why anyone ever paid for an API call. Get a terminal open, because we are going to install something in just a few slides.