Chain-of-thought CoT
A prompting technique where the LLM is asked to think step by step before giving the answer.
Add "let's think step by step" to your prompt, or ask the model to show its intermediate steps. The result is considerably better on reasoning tasks (maths, logic, planning) — the model "tracks" its own thinking.
Modern reasoning models (o1, o3, Claude with thinking mode) do this automatically and invisibly.