Zero, one or few · Few-Shot Prompting: Teach the AI With Examples

A prompt framework is a reusable structure for a request, usually named with an acronym, that acts as a checklist of the parts a good prompt contains.
The AI fills in whatever you leave out with generic guesses, so a framework makes sure you include the parts that change the answer, such as audience, goal and format.
Frameworks make prompts repeatable: you fill the same parts each time, save the prompt as a template and share it with colleagues.
Frameworks are not magic words. The value comes from the detail you put in each part, and a quick, simple question can stay short.
There are well over sixty named frameworks, and most remix the same ingredients: role, task, context, audience, format, constraints and examples.
Zero-shot gives an instruction with no examples, one-shot gives one and few-shot gives several. The terms come from the GPT-3 paper by Brown and colleagues in 2020.

Level	What you give	Typical result
Zero-shot	Instruction only	Most generic output
One-shot	Instruction plus one example	Tone starts to match
Few-shot	Instruction plus several examples	Pattern and style copied

Do this in your own tools, step by step. Ticks are saved on this device.

Pick one task you repeat at work, such as product descriptions, replies to reviews or meeting summaries.
Write the zero-shot version: the instruction with no examples.
Collect three examples of output you already like and paste them into the prompt.
Run both prompts and compare the answers side by side.

What is a prompt framework?

A reusable structure or checklist of the parts a good prompt contains.

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

Zero-shot gives no examples, one-shot gives one and few-shot gives several.

Where do the terms zero-shot, one-shot and few-shot come from?

The GPT-3 paper by Brown and colleagues in 2020.

Adam Welcome to Few-Shot Prompting: Teach the AI With Examples. This course is part of Complete AI Training; you'll find the full library at complete A I training dot com. It takes six short chapters.

Paula This part covers the three ways to give an AI examples: none, one, or several. It is the fastest quality upgrade you can make to any prompt.

Adam A prompt is the request you type into an AI tool such as ChatGPT, Claude or Gemini. A prompt framework is a reusable structure for that request. Think of it as a short checklist of the parts a good prompt contains, usually named with an acronym so it is easy to remember.

Paula Why does a checklist help. The AI fills in whatever you leave out with generic guesses. A framework makes sure you include the parts that change the answer, such as who it is for, what you want to achieve and the format you need.

Adam Frameworks also make prompts repeatable. You fill the same parts each time, save the prompt as a template and share it with colleagues. But they are not magic words. The value comes from the detail you put in each part. A quick, simple question can stay short.

Paula There are well over sixty named frameworks, and most of them remix the same ingredients: role, task, context, audience, format, constraints and examples. Few-shot prompting is different. It is a technique from AI research. Instead of slots to fill, it changes how the AI works through a task, and you can add it on top of any fill-in framework.

Adam So what is few-shot. Zero-shot gives an instruction with no examples. One-shot gives one example. Few-shot gives several. The terms come from the GPT-3 paper by Brown and colleagues in 2020, which showed models learn tasks from examples placed in the prompt.

Paula Here is the difference in practice. A zero-shot product description request gives the most generic output. Add one example of a description you like and the tone tightens. Add three and the AI copies your pattern, your length and your rhythm. Examples are the cheapest quality upgrade available.

Adam This is what zero-shot looks like in the tool. One instruction, no examples. The AI has to guess your style, your length and your audience. Watch how broad the answer comes back.

Paula Now the same writer adds three examples of descriptions they already like. Same task, same tool, different result. The AI copies the pattern from the examples instead of guessing.

Adam The takeaway is simple. Zero-shot gives the most generic output. One example steers it. Several examples teach it. Examples are the cheapest quality upgrade you can make, so reach for them before you rewrite the prompt.

Paula Your turn. Take a task you repeat at work. Write the zero-shot version first, then rewrite it with three examples of output you already like. Compare the two answers side by side.