https://stratechery.com/2025/deepseek-faq/
不懂计术的就不要传小道消息了，我们搞这一行的人每天读deepseek paper.
their most recent DS R1 paper is at https://arxiv.org/pdf/2501.12948
DeepSeek ... implemented cross-GPU communications ... using PTX. They did not use CUDA ... THAT is crazy significant.
H100s were prohibited by the chip ban, but not H800s. Everyone assumed that training leading edge models required more interchip memory bandwidth, but that is exactly what DeepSeek optimized both their model structure and infrastructure around.
Again, just to emphasize this point, all of the decisions DeepSeek made in the design of this model only make sense if you are constrained to the H800; if DeepSeek had access to H100s, they probably would have used a larger training cluster with much fewer optimizations specifically focused on overcoming the lack of bandwidth.
