#cost optimization

reasoning.effort in Practice: Making Reasoning Depth a Tunable Parameter

Same model, adjustable thinking depth. This tutorial covers the reasoning.effort syntax, per-tier scenarios, how it interacts with max_output_tokens and the incomplete status, and the cost consequences of getting it wrong.