Nvidia H100 vs Nvidia RTX 4090

The H100 has 3.3x the memory and 3.3x the bandwidth. The RTX 4090 is typically cheaper.

At a glance

Memory

3.3x more on H100

H100: 80 GB

RTX 4090: 24 GB

Memory bandwidth

3.3x higher on H100

H100: 3.35 TB/s

RTX 4090: 1.01 TB/s

Cheapest on-demand

76% cheaper on RTX 4090

H100: $1.25 /GPU/hr

RTX 4090: $0.30 /GPU/hr

Price comparison

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

Billing	H100	RTX 4090	Difference
On-demand	$3.38 /GPU/hr from $1.25	$0.48 /GPU/hr from $0.30	86% cheaper on RTX 4090
Reserved	$3.14 /GPU/hr from $1.53 (1 mo)	$0.52 /GPU/hr from $0.28 (48 mo)	84% cheaper on RTX 4090
Spot	$2.08 /GPU/hr from $0.03	$0.37 /GPU/hr from $0.16	82% cheaper on RTX 4090

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

H100 vs RTX 4090 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	H100	RTX 4090	Difference
Memory
Capacity	80 GB HBM3 94 GB on NVL	24 GB GDDR6X	3.3x more on H100
Bandwidth	3,350 GB/s 2,000 GB/s on PCIe 3,938 GB/s on NVL	1,010 GB/s	3.3x higher on H100
Bus width	5,120-bit	384-bit	13x wider on H100
Compute · dense
FP8	1,979 TFLOPS	330.3 TFLOPS	6x higher on H100
FP16 / BF16	989.5 TFLOPS	165.2 TFLOPS	6x higher on H100
INT8	1,979 TOPS	660.6 TOPS	3x higher on H100
FP32	67 TFLOPS	82.6 TFLOPS	23% higher on RTX 4090
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	450 W	56% higher on H100
Released	Q3 2022	Q4 2022
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 4090 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 4090
Lium Distributed	$1.25 /GPU/hr On-demand PCIe	$0.30 /GPU/hr On-demand
Vast.ai USA	$1.74 /GPU/hr On-demand SXM	$0.36 /GPU/hr On-demand
GPU.ai UAE	$1.98 /GPU/hr On-demand PCIe	$0.34 /GPU/hr On-demand
Runpod USA	$2.59 /GPU/hr On-demand NVL	$0.34 /GPU/hr On-demand
HyperAI Hong Kong	$1.80 /GPU/hr On-demand	--
EmpirioLabs AI USA	$2.50 /GPU/hr On-demand	--
Koyeb France	$2.50 /GPU/hr On-demand	--
Spheron Singapore	$2.64 /GPU/hr On-demand PCIe	--
Salad Distributed	--	$0.33 /GPU/hr On-demand
Packet·ai USA	--	$0.39 /GPU/hr On-demand
Oblivus USA	--	$0.64 /GPU/hr On-demand PCIe
Runcrate USA	--	$0.66 /GPU/hr On-demand PCIe