Fine-tuning
What is fine-tuning?
Fine-tuning further trains an existing model on your own examples so it adopts a specific format, tone or classification behaviour. It changes how a model responds, not what facts it knows.
When to fine-tune
- Consistent output format at high volume
- Domain-specific classification with plenty of labelled data
- Latency or cost reduction by moving work to a smaller model
When not to
If the problem is 'the model doesn't know our facts', use retrieval-augmented generation instead.
Last reviewed by the gAIcko editorial team.