# Fine-Tuning

> Training an existing model further on task-specific examples or domain data.

- Category: Models & Training
- Stage: Established
- Also called: Model customization, Supervised fine-tuning
- Canonical: https://www.joelcsaji.com/ai-dictionary/fine-tuning

## Plain English

You teach a model a narrower style, format, or behavior after its broad training.

## Example

A company fine-tunes a model to classify support tickets using thousands of labeled historical examples.

## Why it matters

Fine-tuning can improve consistency, but retrieval and prompting may be simpler for knowledge updates.

## Related

- [Distillation](https://www.joelcsaji.com/ai-dictionary/distillation)
- [Synthetic Data](https://www.joelcsaji.com/ai-dictionary/synthetic-data)
- [RAG](https://www.joelcsaji.com/ai-dictionary/retrieval-augmented-generation)

## Sources

- [Google Cloud Generative AI Glossary](https://docs.cloud.google.com/docs/generative-ai/glossary)