OpenAI: o3 Pro

The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently better answers.

Note that BYOK is required for this model. Set up here: https://openrouter.ai/setting...