Share via


ReasoningEffort Enum

o-series models only

Constrains effort on reasoning for reasoning models. Currently supported values are low, medium, and high. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.

Fields

HIGH
LOW
MEDIUM