First B200 rack
Hardware
One rack, four nodes, all pre-sold to waitlist. Interesting: the wall-plug efficiency versus H100 on an FP8 training benchmark we run for burn-in is about 2.1× per watt. Less interesting: the busway upgrade it needed.
Rent the hardware this post is about
Dedicated GPUs by the week or the month, paid in crypto, in three US datacenters.