Skip to content

Model Serving & Economics

Reasoning tokens

Small units counted as part of an AI's behind-the-scenes work toward an answer.

Example

A service reports reasoning-token usage in addition to visible answer text.

Why people use it

It helps account for work an AI system performs beyond the visible reply.

What you'll hear

“Why is the usage higher than the words I can see?”

What this means for you

Check how reasoning usage affects limits, latency and cost.

Can you control it?

Sometimes

Sometimes. Your choices depend on the tool and your access. The settings available to an everyday user may differ from those available to the people running it.

Common questions

Are reasoning tokens always visible to the user?
No. Reporting and visibility depend on the service.
Can they affect cost?
Yes. Depending on the service, this work can count toward usage charges.
Does a larger count prove better reasoning?
No. More intermediate work can still lead to a mistaken answer.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice