Fine-tuning
The process of taking a pre-trained AI model and training it further on a specific dataset to permanently alter its behavior, style, or specific knowledge.
What it is
Fine-tuning adjusts the underlying weights of an AI model. Instead of relying purely on instructions in the prompt (like RAG), you feed the model thousands of examples of the desired input and output. This trains the model to consistently adopt a specific tone, format JSON perfectly, or understand highly niche industry jargon without needing massive prompt context.
When you would use it
You use fine-tuning when prompt engineering and RAG fail to produce consistent formatting or stylistic results, and you have a large dataset of high-quality examples to train on.
Common operations
- Training a model to strictly output data in a proprietary API format.
- Adjusting a model's tone to perfectly match a company's brand voice.
Related terms
Where this is taught
No learning path uses this term yet. Browse the Learning Atlas for guided sequences through related ideas.