DeepSeek released a new AI reasoning model that not only bests OpenAI’s o1-preview in many tests but also shows you how the model is thinking through the problems. It’s an interesting look at how these models use chain-of-thought to do solve these sorts of problems, especially when it comes to spelling strawberry.
Get the full conversation in this week's episode!
Deepseek Reasoning Model
https://www.deepseek.com/
