Thinking and effort
Also called Reasoning, Extended thinking, Effort level
Tokens a model produces before its answer, working a problem through, which tends to help on hard tasks. Effort is the setting that controls how much of this happens. More effort costs more and takes longer.
Effort is a dial to try before changing model. Thinking is billed at the output price, whether you can see it or not, and it isn't a trustworthy record of why the model did something.
Say it to your agent
“Run the codebase search at effort low and the migration design at effort high. Show me usage.output_tokens_details.thinking_tokens for both before we change anything else.”