This was part of the premise of o1 though, no? By encouraging the model to output shorter/longer chains of thought, you can scale model performance (and costs)

This was part of the premise of o1 though, no? By encouraging the model to output shorter/longer chains of thought, you can scale model performance (and costs) down/up at inference time.