Chain-of-thought (CoT) prompting asks the model to work through a problem step by step before it answers. On multi-step tasks such as arithmetic, logic and planning, writing out intermediate steps often improves accuracy, at the cost of longer and slower responses.
Many techniques build on the idea. Self-consistency samples several reasoning paths and takes the majority answer. Least-to-most and Self-Ask break a hard question into smaller ones. Tree of Thoughts explores several branches and backtracks. Program-of-Thought writes the steps as code. Reasoning models now do much of this internally, which changes when explicit CoT is worth it. These guides start with step-by-step prompting for beginners and move on to the advanced variants, with notes on when each one helps.
The library lists all guides with filters for topic, level, category and reading time.