NVIDIA datacenter GPUs
Every A100, H100, H200, B200 and GB200 racked and powered on Earth — the single largest block of purpose-built AI compute, and about 62% of the world's AI accelerator capacity. Quoted at dense BF16: a B200's headline '20 PFLOPS' is FP4 with sparsity, which is 2.25 PFLOPS here.
NVIDIA datacenter GPUs over time
Installed compute in zettaFLOP/s, log scale, 2008 → present. A straight line on this axis means steady exponential growth; the slope is the doubling rate. History before the 2026 anchor is reconstructed — see the methodology.
Epoch AI fleet estimate x NVIDIA share, converted at dense BF16 per H100-equivalent.
~9.3M H100-equivalents. Blackwell (B200/B300/GB200) now carries the majority; Hopper is the long tail.
Overlap treatment: Datacenter parts only. GeForce cards sold into gaming PCs are the separate consumer line, and the TOP500 machines built from these GPUs are the not-summed overlay.
Growth: compounding at 2.60× a year (8.7-month doubling) between published vintages.