AI
Reasoning
A model generating intermediate steps or chains of thought before arriving at an answer, trading latency for accuracy on complex problems.
Reasoning models (o1, o3, Claude with extended thinking) allocate more compute per query. The cost is higher latency and token spend. The benefit is measurable on tasks that require multi-step logic, maths and code, and negligible on tasks that do not. Not all reasoning is visible: some models produce internal chain-of-thought tokens that are not shown to the user.