easyAI Engineering

What is the difference between zero-shot, one-shot and few-shot prompting?

Q: What is the difference between zero-shot, one-shot and few-shot prompting?

**TL;DR.** **Zero-shot** = instruction only; **one-/few-shot** = include worked examples so the model infers the pattern (**in-context learning**, no training). Modern chat models are strong zero-shot — add examples when **format or edge-case behaviour** must be nailed, not by default. zero-shot instruction input → ? <text class='d-text' x='3

191 views

Understand the problem

How in-context examples steer a model without any training.

few-shotzero-shotin-context-learningprompting

Attempt it yourself

Sketch your approach before reading the solution — that's what interviews test.

Stuck? AI Nudge Available

Get a conceptual hint to guide your logic without spoiling the final implementation.

Study the solution

The solution is waiting

Give it an honest attempt first — then compare your thinking with the full walkthrough.

Read the code

Few-shot with hard negatives for a classifier

const prompt = [
  "Classify the ticket as: bug | feature_request | question.",
  "",
  'Ticket: "App crashes when I tap export" → bug',
  'Ticket: "Export to CSV would be great" → feature_request',
  // hard negative: sounds like a bug, is actually a question
  'Ticket: "Is export supposed to include archived items?" → question',
  "",
  "Ticket: " + JSON.stringify(ticketText) + " →",
].join("\n");

Join the discussion

Discussion (0)

No responses yet. Be the first to share what you think.

What are hallucinations and what causes them?

What is a context window and what happens when you exceed it?