Few-shot
Giving an AI a small number of examples in the prompt to show it exactly what you want.
Few-shot prompting means including two to five examples of the task you want completed directly in your prompt, before asking the model to do it for real. The model picks up the pattern from your examples and applies it to the new input. It sits between zero-shot (no examples) and fine-tuning (retraining the model on many examples).
Imagine asking a new colleague to write in your company's specific tone. You could describe it in words (zero-shot), or you could show them three example emails and say "write like this" (few-shot). Most people learn better from examples than from descriptions — and so do AI models.
Few-shot prompting doesn't teach the model anything permanently. The examples only exist for that one conversation. Once the conversation ends, the model retains nothing. It's in-context learning, not training — the model is pattern-matching from your examples in real time, not updating its weights.