A setting for how much thinking a model spends before it answers.
It is like answering at a game-show buzzer or with math scratch paper. Same question, very different odds.
You see it as a setting on reasoning models. Higher effort helps on hard tasks, but it is slower and costs more compute.
Reasoning-model
Reasoning models often offer this “think more” control.
Chain-of-thought
Reasoning effort can make the middle thinking deeper and more detailed.
Inference
Higher reasoning effort usually makes inference slower and more costly.
Token
Higher reasoning effort often uses more tokens, especially on hard tasks.