NVIDIA · Data centre

V100

Rented by 2 providers. 2 exact variants are priced separately, because they are different silicon.

Best published rate

$0.190

per GPU / hour
RunPod · 16 GB

Every published rate

cheapest first · USD per GPU per hour

Provider	What they call it	Their own price, and the arithmetic	Per GPU-hour
RunPod 16 GB	Tesla V100 Tesla V100-PCIE-16GB	$0.1900 per hour = $0.1900 per GPU-hour region Community Cloud · 1 offer Low stock	$0.1900 $1.52 per PFLOP-hr
RunPod SXM2 · 16 GB	V100 SXM2 Tesla V100-SXM2-16GB	$0.2300 per hour = $0.2300 per GPU-hour region Community Cloud · 1 offer Low stock	$0.2300 $1.84 per PFLOP-hr
Azure 16 GB	V100 Standard_NC12s_v3	$6.12 per hour ÷ 2 GPUs = $3.0600 per GPU-hour region US East · 584 offers spot $0.565	$3.0600 $1.52 per PFLOP-hr

The silicon

vendor-published

Variant	Architecture	Memory	FP16 dense	Board power
V100 SXM2	Volta 2017	16 GB HBM2	125 TFLOPS	300 W
V100	Volta 2017	16 GB HBM2	125 TFLOPS	300 W

Vendor-published figures. FP16 throughput is dense tensor throughput with no sparsity assumed; vendors often headline the sparse figure, which is twice as large. Vendor page ↗

What they call it

Providers write the same hardware down differently. Every name below is one this index has seen for this chip; nothing was renamed to make the table tidier.

V100 SXM2Tesla V100V100