llm
Fine-tuning
The process of further training a pre-trained model on a specialized dataset to improve its performance in a specific domain.
What Is Fine-tuning
Fine-tuning is additional training of a base model (GPT, Llama) on your dataset. After fine-tuning, the model better understands your domain, style, and tasks.
When to Use It
- You need a specific response style
- Working with specialized terminology
- Classification in a narrow domain
- RAG doesn’t provide sufficient quality
When NOT to Use It
- You need current data (use RAG instead)
- Limited budget (fine-tuning is more expensive)
- The task can be solved with prompt engineering
- Not enough training data (< 100 examples)
Cost
Fine-tuning requires GPU resources and time. For most business tasks, a combination of RAG + good prompts delivers results faster and cheaper.