Which is cheaper, H200 or RTX 4090?

RTX 4090 is currently cheaper on gpuprice.fyi's headline comparison metric, using USD per GPU per hour.

How is the H200 vs 4090 price comparison calculated?

Each listing is normalized to USD per GPU per hour. The comparison uses the reliable index median when available, then the separately labeled community marketplace median, otherwise the cheapest valid listing is shown with quality tier and billing context.

OCPI·LIVE

H100from $1.95H100from $1.99H100from $1.66H200from $3.59H200from $2.25A100from $1.00A100from $1.99A100from $1.39A100from $1.19B200from $3.74L40Sfrom $0.69RTX 4090from $0.34RTX 5090from $0.65H100from $1.95H100from $1.99H100from $1.66H200from $3.59H200from $2.25A100from $1.00A100from $1.99A100from $1.39A100from $1.19B200from $3.74L40Sfrom $0.69RTX 4090from $0.34RTX 5090from $0.65

Index / Compare

Cloud GPU Price Comparison

H200 vs RTX 4090

H200 NVL vs RTX 4090 — current cloud rental prices normalized to USD per GPU per hour, across providers. RTX 4090 is currently the cheaper option — about 91% less on the headline comparison metric.

H200 NVL

NVIDIA Hopper · NVL

$3.79/GPU/hr

community median · 1 provider
range $2.25–$4.63 · 2 providers
cheapest: Shadeform $2.25

↗ Full H200 NVL prices

RTX 4090

NVIDIA Ada Lovelace · PCIE

$0.34/GPU/hr

community median · 1 provider
range $0.34–$0.69 · 2 providers
cheapest: RunPod $0.34

↗ Full RTX 4090 prices

H200 vs RTX 4090 price table

USD per GPU per hour · live provider data

Metric	H200 NVL	RTX 4090
Index median	—	—
Community marketplace median	$3.79/hr	$0.34/hr
Cheapest current row	$2.25 · Shadeform	$0.34 · RunPod
Provider coverage	2 providers · 17 listings	2 providers · 8 listings
Best use of this page	Use this as a price comparison and sourcing snapshot. Performance, memory size, software support, and availability still need to be checked with the provider.

“

Listed price ≠ transaction price

These are published rates, normalized per GPU per hour and labeled by billing mode. The index median uses eligible reliable-tier providers only. Read the methodology →