Intelligently selects the optimal mode for your task. Starts with 1 model and minimal thinking effort, automatically escalating to more powerful modes when needed.
Minimal
256 tokens