Extended thinking
Also: reasoning tokens, thinking tokens, thinking mode, thinking budget
A mode where the model thinks for longer before answering. The visible 'thinking' is text it wrote, not a window into its mind.
The fuller definition
Extended thinking is a mode in which a reasoning model spends additional computation, producing intermediate thinking tokens before its answer. Some products show a summary of that work, and most charge for it.
Invented probabilities. Always picking the top candidate gives the same wording every run; sampling sometimes picks a lower one. Neither setting checks whether the sentence is true.
A concrete example
Set to think longer, a model works through a scheduling problem step by step before proposing a timetable.
The distinction that matters
Visible thinking is generated text, not a transcript of the model's internal state. It can read plausibly and still be wrong.
Related terms
Reasoning model
A model trained to spend extra work on some answers. It may help with hard tasks. It can still be wrong, and speed and cost vary.
Training & generationChain of thought
Having the model write out its steps before the final answer. It often helps, but the written steps are not proof the answer is right.
Prompting & contextOutput token
A piece of text the model writes. Output is usually priced higher than input.
Where it is taught
Why the same question can produce different answersSource and further reading
Anthropic · Context engineeringThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.