Skip to content
BENIBIX Learn

Chain-of-Thought Prompting

AI🛠️ Working with AIPrompt Writing Techniques

Also known as Chain of Thought, CoT Prompting, Step-by-Step Prompting

You ask a model to show its steps first, then give the answer.

Think of it like

Think of it like a teacher who says, show your work. You get the sum right more often.

Example

Jack asks how much 12 pens cost if a pack of 3 costs $2. The quick answer says $6. With a request to think it through, the model works out 4 packs and says $8.

Why it matters

It is one of the simplest ways to cut errors on hard questions.

Definition

Chain-of-thought prompting asks a model to work through a problem step by step before it gives the answer. Writing out the steps gives the model more tokens to reason with. That helps most on maths, logic and tasks with many parts. A 2022 study showed it clearly. Reasoning models now do this on their own, so the prompt matters less with them.

Example code

text
A pack of 3 pens costs $2. How much do 12 pens cost?
Think it through step by step, then give the final answer on its own line.
The last line asks for the steps first. That one change often fixes maths slips.
Answered at once, the model slips. Asked for steps first, it gets the right total.

Common mistake

Trusting the steps as a true record of how the model got there. The written steps can look right and still hide a slip.

Key takeaways

  1. 1
    Steps before answer. Show the work.
  2. 2
    More room to think. More tokens to reason.
  3. 3
    Built into some models. Reasoning models do it.

Learn it properly