Back to Glossary
AI & MLchain-of-thought

Chain-of-Thought (CoT)

Also known asCoTExtended Thinking
A prompting technique or model-native capability where the LLM produces intermediate reasoning steps before arriving at a final answer, improving accuracy on multi-step problems. Originally a prompting strategy ('think step by step'), CoT is now built directly into reasoning models like o1 and DeepSeek-R1 as an internal process. When combining CoT with structured output, developers should place reasoning fields before answer fields to avoid bypassing the reasoning process.

Related terms

3