Tokens and context window: what they mean when running models locally

Tokens are the chunks of text that a model reads and writes. When you're running models locally, you care about speed and memory rather than a per-token bill — there's no metered API cost. The context window is how much information the model can work with at once. Think of it as the model's working memory for a given conversation or task. Both concepts matter for local AI because they directly affect how fast a model runs and how much of your hardware it consumes.