Chain of thought
Asking an LLM to spell out intermediate reasoning steps before producing the final answer. Improves accuracy on multi-step problems by ~10-30 percentage points on most reasoning benchmarks.
Background
Chain of thought (CoT) became standard practice after a 2022 paper showed that just adding 'let's think step by step' raised math-word-problem accuracy dramatically. Modern reasoning models (OpenAI o-series, DeepSeek R1, Claude with extended thinking) generate hidden CoT by default. Visible CoT is still used heavily in classical prompts that target non-reasoning models.
Tools that use it
- 01→Claude
Anthropic's AI assistant with the Claude family of models (Opus, Sonnet, Haiku), Projects, and Computer Use.
- 02→ChatGPT
OpenAI's flagship AI assistant — GPT-5 chat with web, voice, image generation, code interpreter, and custom GPTs.
- 03→DeepSeek Chat
DeepSeek's web chat for DeepSeek-V3 and reasoning models, with open-weights releases.