Few-Shot

Also known as: Few-Shot Learning, N-Shot

Few-shot prompting includes a handful of input–output demonstrations in the prompt so the model infers the pattern. It sits between zero-shot (no examples) and fine-tuning (weight updates).

Use few-shot when the task format is unusual or zero-shot instructions are ambiguous. Keep examples consistent and representative of edge cases you care about.

← Back to AI Glossary