keepUpAI
Start free

AI glossary

What is Fine-Tuning?

Further training a pretrained model on your own examples to change its behavior, style or performance on a narrow task.

Fine-tuning is best for consistent format and tone, or for distilling a big model into a cheaper one. It's a poor way to add changing facts, which is what RAG is for. LoRA makes fine-tuning cheaper by training small adapter matrices.

▶Practice it hands-onPrompting vs RAG vs Fine-Tuning

Related terms

LoRA (Low-Rank Adaptation)RAG (Retrieval-Augmented Generation)

← All 43 AI terms