Few-shot prompting means giving the AI examples of what you want before asking it to do the task. It's the single most reliable technique for getting consistent, formatted outputs.
Zero-Shot vs Few-Shot
Zero-shot means you just describe what you want. Few-shot means you show examples first, then ask.
Here's a zero-shot prompt:
text
Classify the sentiment of this review as positive, negative, or neutral:
"The battery life is incredible but the camera is disappointing."
Here's the same task as a few-shot prompt:
text
Classify the sentiment of each review:
Review: "Absolutely love this product, best purchase ever!"
Sentiment: positive
Review: "Broke after two days. Complete waste of money."
Sentiment: negative
Review: "It works fine. Nothing special."
Sentiment: neutral
Review: "The battery life is incredible but the camera is disappointing."
Sentiment:
The few-shot version is more reliable because the model can see exactly what format and logic you expect.
Structuring Your Examples
The best few-shot prompts follow this pattern:
text
[Instruction describing the task]
[Example input 1]
[Example output 1]
[Example input 2]
[Example output 2]
[Actual input]
[Let the model complete]
How Many Examples?
•1-2 examples work for simple formatting tasks
•3-5 examples work for classification and extraction
•5+ rarely needed — if you need more, the task might be too complex for few-shot alone
Pro Tip: Include Edge Cases
If your task has tricky edge cases, make one of your examples cover it. The model learns from the patterns you show it.