A model is the trained engine that turns your input into an answer, recommendation, or action plan. It’s the capability layer inside a larger system: ChatGPT is an app, Claude Code is a harness (software wrapped around a model), and inside those products one or more models do the actual reasoning while the product decides the tools and rules around them. A large language model is a model trained around language; inference is the model doing the work in the moment.
Hiring people works the same way. Two hires can both be smart and still differ: one is fast with routine admin, one is better at strategy, one is careful with legal language. Models are the same, all capable but good at different work.
Why you care
You rarely need a new model, just the right model tier, context, and instructions. Ask a strong model to explore a messy new problem; once the work becomes a repeatable skill or process, a faster, cheaper one can carry it. When someone says “we trained it on our proposals,” they usually mean they gave it examples in context: useful, and different from fine-tuning, where training changes the model itself.