Last updated
Chain-of-Thought Prompting
Technique: For problems that need reasoning (maths, logic, multi-step decisions), ask the AI to think step by step before answering. Working through the steps makes it far more accurate.
The magic phrase
A shop had 120 apples, sold 35, then received 50 more. How many now?
A shop had 120 apples, sold 35, then received 50 more. How many now? Think step by step, then give the final answer.
Start: 120. After selling 35: 120 - 35 = 85. After receiving 50: 85 + 50 = 135.
Final answer: 135 apples.
Why it works
Forcing the model to lay out intermediate steps reduces careless errors, just like showing your working in a maths exam. It also lets you spot exactly where any mistake happened.
When to use it
- Arithmetic or word problems
- Logic puzzles and comparisons
- Decisions with several factors to weigh
💡 Tip: if you only want the final answer, add "then give only the final result on the last line."