Nvidia A100 vs Nvidia B200

The B200 is newer, with 2.2x the memory and 4x the bandwidth. The A100 is typically cheaper.

At a glance

Memory

2.2x more on B200

A100: 80 GB

B200: 180 GB

Memory bandwidth

4x higher on B200

A100: 1.94 TB/s

B200: 7.70 TB/s

Cheapest on-demand

89% cheaper on A100

A100: $0.43 /GPU/hr

B200: $3.75 /GPU/hr

Price comparison

Median price per GPU per hour by billing type, with the cheapest listing under it.

Billing	A100	B200	Difference
On-demand	$1.88 /GPU/hr from $0.43	$7.99 /GPU/hr from $3.75	76% cheaper on A100
Reserved	$1.67 /GPU/hr from $0.66 (24 mo)	$5.62 /GPU/hr from $3.75	70% cheaper on A100
Spot	$1.10 /GPU/hr from $0.68	$4.26 /GPU/hr from $3.94	74% cheaper on A100

Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. Verify before provisioning. More on how we price.

Price history

A100 vs B200 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	A100	B200	Difference
Memory
Capacity	80 GB HBM2e 40 GB on SXM, PCIe	180 GB HBM3e	2.2x more on B200
Bandwidth	1,935 GB/s 2,039 GB/s on SXM 1,555 GB/s on SXM, PCIe	7,700 GB/s	4x higher on B200
Bus width	5,120-bit
Compute · dense
FP4	Not supported	9,000 TFLOPS	B200 only
FP8	Not supported	4,500 TFLOPS	B200 only
FP16 / BF16	312 TFLOPS	2,250 TFLOPS	7.2x higher on B200
INT8	624 TOPS	4,500 TOPS	7.2x higher on B200
FP32	19.5 TFLOPS	75 TFLOPS	3.8x higher on B200
FP64	9.7 TFLOPS	37 TFLOPS	3.8x higher on B200
Platform
Interconnect	NVLink 600 GB/s	NVLink 1,800 GB/s	3x higher on B200
Board power	300 W	1,000 W	3.3x higher on B200
Released	Q2 2020	Q1 2024	4 years newer on B200
Precision
Supported formats	FP16BF16TF32FP32FP64INT4INT8	FP4FP6FP8FP16BF16TF32FP32FP64INT8

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.

Source: A100 datasheet and B200 datasheet.

Where to rent them

Providers that rent these GPUs, with each one's cheapest rate. Providers that rent both are listed first.

Provider	A100	B200
GPU.ai UAE	$0.43 /GPU/hr On-demand	$7.99 /GPU/hr On-demand SXM
Packet·ai USA	$1.43 /GPU/hr On-demand	$3.75 /GPU/hr On-demand
Runpod USA	$1.19 /GPU/hr On-demand PCIe	$7.99 /GPU/hr On-demand SXM
Spheron Singapore	$1.43 /GPU/hr On-demand PCIe	$11.06 /GPU/hr On-demand SXM
Jarvislabs India	$0.89 /GPU/hr On-demand	--
Thunder Compute USA	$1.09 /GPU/hr On-demand	--
Hyperstack UK	$1.35 /GPU/hr On-demand PCIe	--
Massed Compute USA	$1.38 /GPU/hr On-demand SXM	--
Lium Distributed	--	$7.72 /GPU/hr On-demand SXM
Verda Finland	--	$7.87 /GPU/hr On-demand SXM
EmpirioLabs AI USA	--	$15.50 /GPU/hr On-demand
GPUaaS USA	--	$7.00 /GPU/hr Custom