POWERING THE NEXT AGE OF INTELLIGENCE-RTX PRO 6000 BLACKWELL 96GB

NVIDIA H100 (80GB HBM3)

80GB HBM3
Memory
3,350GB/s
Memory bandwidth
Q3 2022
Released

Price history

On-demandReservedSpot
USD / GPU / HR

What it costs to rent a NVIDIA H100

The NVIDIA H100 is a Hopper-architecture fine-tuning NVIDIA GPU with 80GB of HBM3 memory and 3,350 GB/s of memory bandwidth, released Q3 2022. It is built for large-model training and high-throughput inference.

Renting one currently costs between $1.00 and $12.29 per GPU-hour across the 37 cloud providers we track, with a median of $3.39. The cheapest live rate is $1.00/hr from Vast.ai (Spot), and hosting one pays about $2.30/hr on RunPod while it is running. All rates below are normalised to a single GPU per hour so providers with multi-GPU nodes compare like-for-like.

Where to rent the H100

How we rank

The median is one provider, one vote: each provider contributes the median of its own on-demand listings, then the headline is the median across providers — so no single marketplace or hyperscaler sets the price alone. Below two providers we show the range instead.

Default order is relevance, weighed the way a buyer would: price per GPU first, lightly blended with total node cost; on-demand preferred over spot and reserved when prices are close; established clouds edged above marketplaces; in-stock, recently-checked offers above stale ones; and no single provider is allowed to crowd the top. Click a column to sort by the raw numbers. No paid placements — ranking is never sold.

ProviderTierConfigTotal VRAMBilling$ / GPU / hrTotal / hrAvailabilityGo to provider
Vast.aiGPU cloud1× H100 SXM80GBSpot$1.00$1.00In stockRent
Clore.aiGPU cloud2× H100 NVL160GBOn-demand$1.15$2.29In stockRent
NosanaGPU cloud1× H10080GBCommunity$1.02$1.02UnknownRent
Google CloudHyperscaler1× H10080GBSpot$1.09$1.09UnknownRent
Vast.aiGPU cloud4× H100 NVL320GBSpot$1.00$4.00In stockRent
AzureHyperscaler1× H10080GBSpot$1.29$1.29UnknownRent
NosanaGPU cloud1× H10080GBOn-demand$1.36$1.36UnknownRent
AzureHyperscaler2× H100160GBSpot$1.29$2.58UnknownRent

Earn by hosting the H100

How this is worked out

“Renters pay” is the median live rate for this GPU on that marketplace — the median, not the floor, because the cheapest listing is one host undercutting everyone. “You earn” is that figure less the platform’s cut.

“Per day” is a ceiling: 24 hours, every hour rented, which no host achieves — scale it by the utilisation you expect. A card sitting idle earns nothing.Fees marked as published come from the platform’s terms; the rest are our estimates, labelled on each row.

PlatformRenters payFeeYou earnPer dayAvailabilityGo to platform
RunPod$3.29/hr730%$2.30/hr$55/dayUnknownHost
Akash$2.56/hr120%$2.05/hr$49/dayUnknownHost
Nosana$1.52/hr310%$1.36/hr$33/dayUnknownHost
Vast.ai$1.67/hr3125%$1.25/hr$30/dayUnknownHost
Clore.ai$1.15/hr210%$1.03/hr$25/dayUnknownHost

Technical specifications

Throughput
FP6426 TFLOPS
FP3251 TFLOPS
TF32378 TFLOPS
FP16/BF16756 TFLOPS
Silicon & memory
Memory
80 GB
Memory bandwidth
2,040 GB/s
TDP
700 W
Process
4 nm
Die size
814 mm²
Transistors
80.0 B
Foundry
TSMC
Released
20 Sep 2022
TrainingInference
Datasheet
ArchitectureHopper
Memory80 GB HBM3 (SXM)
Bandwidth3.35 TB/s
FP8 Tensor3,958 TFLOPS (sparse)
NVLink900 GB/s
MIGup to 7 instances
Powerup to 700 W (SXM), 350 W (PCIe)
Class
Fine-tuning
Cloud providers
37 tracked
Price range
$1.00 – $12.29 / GPU / hr

SNAPSHOT 2026-08-23T22:42:15.889486+00:00 · BACK TO THE FULL INDEX