alt Hacker NewsIt seems particularly poor compared to larger deployments of other models like Kimi K3.1 Pro.
I've heard numbers between 10k-38k tokens per second per GB300 GPU.
It seems particularly poor compared to larger deployments of other models like Kimi K3.1 Pro.
I've heard numbers between 10k-38k tokens per second per GB300 GPU.
GB300 is a full rack of 72 GPUs and costs ~10x more than two DGXs in the article.