Nvidia B100 vs Nvidia B200

Two Nvidia data center GPUs on the Blackwell architecture: up to 192 GB HBM3e against 180 GB HBM3e and memory bandwidth within 5%.

At a glance

Updated 6 minutes ago

Nvidia B100

Nvidia B200

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

Billing	B100	B200
On-demand	--	$3.75
Spot	No spot listings	$3.12
Reserved	No reserved listings	$3.75
Median	--	$6.83

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	B200	Difference
Memory
Capacity	192 GB HBM3e	180 GB HBM3e	1.1x more on B100
Bandwidth	8,000 GB/s	7,700 GB/s	Within 5%
Compute · dense
FP4	7,000 TFLOPS	9,000 TFLOPS	1.3x higher on B200
FP8	3,500 TFLOPS	4,500 TFLOPS	1.3x higher on B200
FP16 / BF16	1,750 TFLOPS	2,250 TFLOPS	1.3x higher on B200
INT8	3,500 TOPS	4,500 TOPS	1.3x higher on B200
FP32	60 TFLOPS	75 TFLOPS	1.2x higher on B200
FP64	30 TFLOPS	37 TFLOPS	1.2x higher on B200
Platform
Interconnect	NVLink 1,800 GB/s	NVLink 1,800 GB/s	Identical
Released	Q4 2024	Q1 2024	7 months newer on B100
Precision
Supported formats	FP4FP6FP8FP16BF16TF32FP32FP64INT8	FP4FP6FP8FP16BF16TF32FP32FP64INT8

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.