Nvidia A40 vs Nvidia RTX 3070

A data center GPU against a consumer card on the Ampere architecture: 48 GB GDDR6 against 8 GB GDDR6 and 55% more memory bandwidth on the A40.

At a glance

Updated 7 minutes ago

Nvidia A40

Nvidia RTX 3070

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

Billing	A40	RTX 3070	Difference
On-demand	$0.49	$0.10	4.9x more expensive on A40
Spot	No spot listings	$0.04
Reserved	$0.61	No reserved listings
Median	$0.77	$0.10	7.4x more expensive on A40

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	A40	RTX 3070	Difference
Memory
Capacity	48 GB GDDR6	8 GB GDDR6	6x more on A40
Bandwidth	696 GB/s	448 GB/s	1.6x higher on A40
Bus width	384-bit	256-bit	1.5x wider on A40
Compute · dense
FP16 / BF16	149.7 TFLOPS	40.6 TFLOPS	3.7x higher on A40
INT8	299.3 TOPS	162.6 TOPS	1.8x higher on A40
FP32	37.4 TFLOPS	20.3 TFLOPS	1.8x higher on A40
Platform
Interconnect	PCIe only	PCIe only
Board power	300 W	220 W	1.4x higher on A40
Released	Q4 2020	Q4 2020
Precision
Supported formats	FP16BF16TF32FP32INT4INT8	FP16BF16TF32FP32INT4INT8

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.