A reasoning model trained with reinforcement learning, on par with OpenAI o1
Want to make some of these yourself?
