Nvidia B100 vs Nvidia P100

Two Nvidia data center GPUs, one Blackwell and one Pascal: up to 192 GB HBM3e against 16 GB HBM2 and 11x the memory bandwidth on the B100.

At a glance

Updated 18 minutes ago

Nvidia B100

Nvidia P100

Cheapest listed price per GPU per hour by billing type, and the median across providers.

Billing	B100	P100
On-demand	--	$0.09
Spot	No spot listings	$0.08
Reserved	No reserved listings	$0.12
Median	--	$1.42

Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Spec	B100	P100	Difference
Memory
Capacity	192 GB HBM3e	16 GB HBM2	12x more on B100
Bandwidth	8,000 GB/s	732 GB/s	11x higher on B100
Bus width		4,096-bit
Compute · dense
FP4	7,000 TFLOPS	Not supported	B100 only
FP8	3,500 TFLOPS	Not supported	B100 only
FP16 / BF16	1,750 TFLOPS	18.7 TFLOPS	94x higher on B100
INT8	3,500 TOPS	Not supported	B100 only
FP32	60 TFLOPS	9.3 TFLOPS	6.5x higher on B100
FP64	30 TFLOPS	4.7 TFLOPS	6.4x higher on B100
Platform
Interconnect	NVLink 1,800 GB/s	NVLink 160 GB/s	11x higher on B100
Released	Q4 2024	Q2 2016	8 years newer on B100
Precision
Supported formats	FP4FP6FP8FP16BF16TF32FP32FP64INT8	FP16FP32FP64

Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.

Compute: Vendor peak figures, dense and per GPU. Not measured throughput.

Providers listing these GPUs, those with both first, so you can switch without moving clouds.