---
Title: "What Is Fine-Tuning in AI? Fine-Tuning vs RAG"
Url: "https://computer.io/glossary/fine-tuning"
Topic: "AI foundations"
Description: "Fine-tuning further trains a pre-trained AI model on specific examples to adapt its behavior. When to fine-tune, how it compares with RAG, and costs."
---

# What is fine-tuning in AI?

Fine-tuning is the process of further training a pre-trained AI model on a smaller, task-specific set of examples to adapt its behavior, style or skills. It is useful for consistent formats and specialized tasks, while RAG is usually better for keeping facts current.

## Fine-tuning vs RAG

|  | Fine-tuning | RAG |
| --- | --- | --- |
| Changes | Model behavior | Information given |
| Fresh data | Needs retraining | Immediate |
| Citations | No | Yes |
| Best for | Style, format, skills | Facts, policies |

## When to fine-tune

Consistent output formats, domain language and classification tasks where prompts alone fall short.

## FAQ

### Is fine-tuning expensive?

It requires training data and compute, though costs have fallen.

### Can fine-tuning reduce hallucinations?

It can improve behavior but doesn't keep facts up to date.

### Should you fine-tune for customer support?

Usually start with RAG and prompts; fine-tune only if needed.

## Related terms

- [Retrieval-augmented generation (RAG)](https://computer.io/glossary/retrieval-augmented-generation)
- [Large language model (LLM)](https://computer.io/glossary/large-language-model)
- [Prompt engineering](https://computer.io/glossary/prompt-engineering)
- [Should you use RAG or fine-tuning for enterprise AI?](https://computer.io/answers/should-you-use-rag-or-fine-tuning)

## How Computer helps

- [How computer works](https://computer.io/how-computer-works)

## Further reading

- [Rag vs knowledge graph ai](https://devrev.ai/blog/rag-vs-knowledge-graph-ai)