# Fine-Tuning

> More training on a model that is done, so it gets good at one job.

**Field:** Artificial Intelligence  
**Category:** LLMs  
**Also known as:** Finetuning, Model Fine-Tuning

## Think of it like

Think of it like a cook who knows how to cook. Then they learn the dishes of one shop.

## Definition

Fine-tuning is more training on a model that is already trained, using a small set of examples for one job. The model keeps what it knew and gets better at that job or style. A few hundred to a few thousand good examples are often enough. It is much cheaper than training from scratch. It changes the model's weights, so the new skill is built in.

## Example

Lucy's support team fine-tunes a model on 2,000 past replies. Now it answers in the firm's tone and knows its product names.

## Why it matters

It is how one general model becomes many special ones. You do not pay for a whole new model.

## Common mistake

Using it to add new facts. Fine-tuning is good at style and format. For facts that change, giving the model the right papers at question time works better.

## Three principles

- **Starts from a trained model.** Not from scratch.
- **Small, focused data.** Hundreds to thousands of examples.
- **Changes the weights.** The skill is built in.

## Related terms

- [Pretraining](https://learn.benibix.com/term/ai/pretraining)
- [Instruction Tuning](https://learn.benibix.com/term/ai/instruction-tuning)
- [Low-Rank Adaptation (LoRA)](https://learn.benibix.com/term/ai/low-rank-adaptation)
- [Transfer Learning](https://learn.benibix.com/term/ai/transfer-learning)
- [Few-Shot Prompting](https://learn.benibix.com/term/ai/few-shot-prompting)
- [Retrieval-Augmented Generation (RAG)](https://learn.benibix.com/term/ai/retrieval-augmented-generation)
- [Parameter-Efficient Fine-Tuning (PEFT)](https://learn.benibix.com/term/ai/parameter-efficient-fine-tuning)

## Told apart from

- [Pretraining vs Fine-Tuning](https://learn.benibix.com/compare/pretraining-vs-fine-tuning): People often think fine-tuning is where a model learns what it knows. Most of its knowledge comes from pretraining, the huge first round on trillions of tokens. Fine-tuning is a small round on a few thousand examples. It shapes style and skill for one job. It is cheap and quick, but it adds few new facts.
- [Few-Shot Prompting vs Fine-Tuning](https://learn.benibix.com/compare/few-shot-prompting-vs-fine-tuning): People often think that examples in a prompt train the model. Few-shot prompting only steers one request. The examples must be sent again each time. Fine-tuning changes the model's weights with many examples, so the skill stays. Few-shot is free to try in a minute. Fine-tuning takes data, time and money.
- [RAG vs Fine-Tuning](https://learn.benibix.com/compare/retrieval-augmented-generation-vs-fine-tuning): People often think fine-tuning is how you teach a model your own documents. Fine-tuning changes how a model writes, and it is poor at adding facts that change. RAG leaves the model alone and hands it the right text at question time, so the facts stay current and can be traced. Use RAG for knowledge. Use fine-tuning for style and format.
- [Hyperparameter Tuning vs Fine-Tuning](https://learn.benibix.com/compare/hyperparameter-tuning-vs-fine-tuning): People think both kinds of tuning change the model, and only one does. Hyperparameter tuning picks the settings for training, like the learning rate, by training many times. Fine-tuning takes a model that is already trained and trains it more on new data. The first picks the recipe. The second keeps cooking.

## Taught in

- [Fine-Tuning and LoRA](https://learn.benibix.com/learn/ai-llms/fine-tuning-and-lora): LLMs, 9/15

![A general model and a small set of examples for one job go into more training. Out comes a tuned model that is good at that job.](https://learn.benibix.com/diagrams/llm-fine-tuning.svg)

---

Source: https://learn.benibix.com/term/ai/fine-tuning
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-18
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
