Back to the wiki

Fine-tuning

Turning a generalist model into your specialist.

The analogy

A freshly graduated doctor knows a bit of everything. A cardiology residency turns them into a specialist: same brain, extra training in one specific area. Fine-tuning is that residency — for an AI model.

In detail

It means continuing to train an existing model on your own set of examples (hundreds or thousands) to adjust its behavior: tone, format, domain vocabulary. Unlike RAG, which adds knowledge the model can consult, fine-tuning changes the model's weights. It's useful when you need a very consistent style or task.

An example

An insurance company fine-tunes a model on thousands of real answers from its customer-service team. The result responds with the company's exact tone, format and terminology.

Related concepts