NLP & Language · intermediate · concept 120 of 311 · Updated Nov 4, 2026
Chain-of-Thought Reasoning
A prompting technique where the model 'shows its work' by reasoning step-by-step before answering. Dramatically improves performance on math, logic, and complex reasoning tasks.
Top AI contributors on this
- Jakub PachockiOpenAI
- Shengjia ZhaoMeta Superintelligence Labs
- Mark ChenOpenAI
Key terms
Step-by-step reasoningScratchpadTree-of-thoughtSelf-consistency
Learn these first
Videos
▶ Stanford CS25: V5 I Large Language Model Reasoning, Denny Zhou of Google Deepmind ↗
Stanford Online · YouTube
▶ How do thinking and reasoning models work? ↗
Google for Developers · YouTube
Guides and articles
Why We Think ↗
Lilian Weng
GitHub ↗
Prompting Guide
Courses, papers, and more