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.