- Glossary
- Reasoning tokens
What are reasoning tokens?
Reasoning tokens are tokens a model generates while working through a problem before it writes its final answer. They are billed as output tokens even when the provider does not show the reasoning text. Higher reasoning effort usually improves hard tasks and raises both cost and response time.
Also called: Thinking tokens, Extended thinking · Last updated
Controlling effort
Many reasoning models accept an effort setting from low to high. Use low effort for simple tasks and reserve high effort for problems that need multi-step work.
Reading the usage
API responses report reasoning tokens as a detail of completion tokens, so you can see how much of the output cost came from reasoning.