PetaPrice

Hopper vs Blackwell

NVIDIA's two current datacenter generations, compared on specs and on what they cost to rent right now.

B200 SXM vs H100 SXM

Right now the B200 SXM is the cheapest per unit of dense FP8 compute among these GPUs: $0.71 per PFLOP-hour at $3.20 per GPU-hour.

In short

Choose Hopper (H100, H200) if

  • you want the most providers to choose from
  • your model fits in 80–141 GB per GPU
  • a low price per hour matters more than time to result
  • you need FP64: Blackwell has far less of it

Choose Blackwell (B200, B300) if

  • your software uses FP8 or FP4 and keeps the GPU busy
  • you need 180–288 GB of memory per GPU
  • you want training runs to finish sooner
  • you serve large models and can quantize to FP4

Specs and prices

H100 SXMH200 SXMB200 SXMB300 SXM
ArchitectureHopperHopperBlackwellBlackwell Ultra
Memory80 GB HBM3141 GB HBM3e180 GB HBM3e288 GB HBM3e
Memory bandwidth3,350 GB/s4,800 GB/s7,700 GB/s8,000 GB/s
FP4 (dense)——9 PF13.5 PF
FP8 (dense)1.98 PF1.98 PF4.5 PF4.5 PF
FP16/BF16 (dense)0.989 PF0.989 PF2.25 PF2.25 PF
FP64 (dense)0.067 PF0.067 PF0.037 PF0.0013 PF
Power700 W700 W1,000 W1,100 W
Lowest on-demand$1.82$2.25$3.20$6.60
$ per PFLOP·h (FP8)$0.92$1.14$0.71$1.47
Providers22171112

Every Hopper and Blackwell datacenter GPU

With its lowest on-demand price and cost per PFLOP-hour of dense FP8.

  1. 1B200 SXM$3.20/h$0.71/PFLOP·h
  2. 2H100 SXM$1.82/h$0.92/PFLOP·h
  3. 3GH200$1.99/h$1.01/PFLOP·h
  4. 4H100 PCIe$1.66/h$1.10/PFLOP·h
  5. 5H200 SXM$2.25/h$1.14/PFLOP·h
  6. 6B300 SXM$6.60/h$1.47/PFLOP·h
  7. 7H100 NVL$2.59/h$1.55/PFLOP·h
  8. 8GB300$9.72/h$1.94/PFLOP·h
  9. 9H200 NVL$3.62/h$2.17/PFLOP·h
  10. 10GB200$16.0/h$3.20/PFLOP·h

Compare two GPUs directly

Hopper vs Blackwell: questions

Is Blackwell worth it over Hopper?

The B200 SXM has 2.3× the dense FP8 compute of the H100 SXM at 1.8× the lowest hourly price right now, so it is 22% cheaper per unit of FP8 compute. It also has 180 GB of memory versus 80 GB. If your software uses FP8 or FP4 and keeps the GPU busy, that's what you pay for; if availability or FP64 matters more, Hopper is the safer choice.

What does Blackwell's two-die design mean?

The B200 joins two large dies with a 10 TB/s link so that software sees a single GPU. It gets around the size limit of a single chip.

How much more power does Blackwell draw?

The B200 SXM is rated at 1000 W, the H100 SXM at 700 W. When you rent, power is part of the hourly price.

Is the H200 a Blackwell GPU?

No. The H200 is a Hopper GPU with more and faster memory than the H100 (141 GB of HBM3e); its compute is the same.