rigami.
Glossary
/// Definition

Fine-tuning

Fine-tuning adapts a pretrained LLM to a specific task or style by continuing training on a smaller labelled dataset.

In depth

Fine-tuning updates the weights of a pretrained model using a targeted dataset. In 2026 the frontier fine-tuning options are LoRA / QLoRA adapters for open-weight models and provider-hosted fine-tuning for OpenAI and Google. RAG is usually a better first move than fine-tuning.

Related terms