#459 – DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters

10405.429 - 10428.199 Dylan Patel

It's even less prevalent. So it's... The remarkable thing about these reasoning results, and especially the DeepSeq R1 paper, is this result that they call DeepSeq R1-0, which is they took one of these pre-trained models, they took DeepSeq V3 base, and then they do this reinforcement learning optimization on verifiable questions or verifiable rewards for a lot of questions and a lot of training.