Nvidia H100 vs Nvidia RTX 4000 Ada

The H100 has 4x the memory and 9.3x the bandwidth. The RTX 4000 Ada is cheaper right now.

At a glance

Memory

4x more on H100

H100: 80 GB

RTX 4000 Ada: 20 GB

Memory bandwidth

9.3x higher on H100

H100: 3,350 GB/s

RTX 4000 Ada: 360 GB/s

Cheapest on-demand

71% cheaper on RTX 4000 Ada

H100: $1.80 /GPU/hr

RTX 4000 Ada: $0.52 /GPU/hr

Price comparison

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

Billing	H100	RTX 4000 Ada
On-demand	$3.71 /GPU/hr from $1.80	-- /GPU/hr from $0.52
Reserved	$3.29 /GPU/hr from $1.53 (1 mo)	On request
Spot	$2.31 /GPU/hr from $1.29	No spot listings

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. No median for on-demand on the RTX 4000 Ada: we only show one when at least 3 providers list the GPU on that billing type. Verify before provisioning. More on how we price.

Price history

H100 vs RTX 4000 Ada price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	H100	RTX 4000 Ada	Difference
Memory
Capacity	80 GB HBM3 94 GB on NVL	20 GB GDDR6	4x more on H100
Bandwidth	3,350 GB/s 2,000 GB/s on PCIe 3,938 GB/s on NVL	360 GB/s	9.3x higher on H100
Bus width	5,120-bit	160-bit	32x wider on H100
Compute · dense
FP8	1,979 TFLOPS	213.8 TFLOPS	9.3x higher on H100
FP16 / BF16	989.5 TFLOPS	106.9 TFLOPS	9.3x higher on H100
INT8	1,979 TOPS
FP32	67 TFLOPS	26.7 TFLOPS	2.5x higher on H100
FP64	34 TFLOPS	Not supported	H100 only
Platform
Interconnect	NVLink 900 GB/s NVLink 600 GB/s on PCIe, NVL	PCIe only	NVLink on H100 only
Board power	700 W	130 W	5.4x higher on H100
Released	Q3 2022	Q3 2023	10 months newer on RTX 4000 Ada
Precision
Supported formats	FP8FP16BF16TF32FP32FP64INT8	FP8FP16BF16TF32FP32INT4INT8

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: H100 datasheet and RTX 4000 Ada datasheet.

Where to rent them

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

Provider	H100	RTX 4000 Ada
Runpod USA	$3.19 /GPU/hr On-demand NVL	On request Reserved
Sesterce France	$2.75 /GPU/hr On-demand	$0.87 /GPU/hr On-demand
HyperAI Hong Kong	$1.80 /GPU/hr On-demand	--
GPU.ai UAE	$1.98 /GPU/hr On-demand PCIe	--
Lium Distributed	$2.10 /GPU/hr On-demand SXM	--
Massed Compute USA	$2.73 /GPU/hr On-demand	--
Runcrate USA	$3.25 /GPU/hr On-demand SXM	--
Lambda Labs USA	$3.29 /GPU/hr On-demand PCIe	--
Akamai Cloud USA	--	$0.52 /GPU/hr On-demand
DigitalOcean USA	--	$0.76 /GPU/hr On-demand