456
Dictionary

Chain of Thought

Prompting an LLM to reason step-by-step before answering, often improving accuracy on hard tasks.

1 min readupdated 2026-06-22

/ quick answer

Chain-of-Thought (CoT) is a prompting technique where the model writes intermediate reasoning steps before the final answer. Newer 'reasoning' models (o-series, Claude thinking, Gemini thinking) run CoT internally. Prompting an LLM to reason step-by-step before answering, often improving accuracy on hard tasks.

Prompting an LLM to reason step-by-step before answering, often improving accuracy on hard tasks. Chain-of-Thought (CoT) is a prompting technique where the model writes intermediate reasoning steps before the final answer. Newer 'reasoning' models (o-series, Claude thinking, Gemini thinking) run CoT internally. In practice: Instead of 'What is 17 * 24?', prompt 'Think step by step, then answer. What is 17 * 24?' — the model breaks it into partial products and gets the correct 408. This dictionary node is part of the Onexial knowledge graph and links to related concepts, workflows and tools below.
Definition
Chain-of-Thought (CoT) is a prompting technique where the model writes intermediate reasoning steps before the final answer. Newer 'reasoning' models (o-series, Claude thinking, Gemini thinking) run CoT internally.
Example
Instead of 'What is 17 * 24?', prompt 'Think step by step, then answer. What is 17 * 24?' — the model breaks it into partial products and gets the correct 408.
Related Workflows
Related Prompts
/ frequently asked

What is Chain of Thought?

Chain-of-Thought (CoT) is a prompting technique where the model writes intermediate reasoning steps before the final answer. Newer 'reasoning' models (o-series, Claude thinking, Gemini thinking) run CoT internally.

What is an example of Chain of Thought?

Instead of 'What is 17 * 24?', prompt 'Think step by step, then answer. What is 17 * 24?' — the model breaks it into partial products and gets the correct 408.

Why does Chain of Thought matter for AI and automation?

Prompting an LLM to reason step-by-step before answering, often improving accuracy on hard tasks. It connects to the workflows, prompts and tool stacks linked on this page, so you can move from definition to execution without leaving Onexial.