We match your deposits, up to $150 in free credits. Ends October 1.Learn more  

gpu catalog: h100 nvl

Rent H100 NVL from $2.37/hr

The inference-tuned H100 variant: 94 GB of HBM3 and higher bandwidth than the SXM part, purpose-built for serving LLMs in PCIe servers.

Live platform rates

26 GPU types online
GPUVRAMRegionsAvailabilityPrice / GPU
H100 NVL94G3 regionsavailable$3.19/hrlaunch

The inference-tuned H100 variant: 94 GB of HBM3 and higher bandwidth than the SXM part, purpose-built for serving LLMs in PCIe servers.

VRAM94 GB HBM3
Memory bandwidth3.9 TB/s
FP16 tensor835 TFLOPS
ArchitectureHopper
InterconnectPCIe 5.0 + NVLink bridge
Board power400 W
Released2023
Regionsus-central, us-east, us-west
H100 NVLcommunity94G2 regionsrunning low$2.37/hrlaunch
L40S48G4 regionshigh availability$0.80/hrlaunch
RTX PRO 600096G4 regionshigh availability$1.99/hrlaunch
RTX 6000 ADA48G2 regionshigh availability$0.65/hrlaunch
RTX 5090community32G8 regionshigh availability$0.35/hrlaunch
L4048G2 regionshigh availability$0.78/hrlaunch
RTX 4090community24G8 regionshigh availability$0.27/hrlaunch
RTX A600048G2 regionshigh availability$0.45/hrlaunch
A10080G6 regionshigh availability$1.07/hrlaunch
RTX PRO 450032G2 regionshigh availability$0.72/hrlaunch
H100 SXM80G5 regionshigh availability$3.49/hrlaunch
A100community80G5 regionshigh availability$0.88/hrlaunch
RTX 3090community24G6 regionshigh availability$0.12/hrlaunch
H200 SXM141G6 regionshigh availability$3.59/hrlaunch
RTX 409024G3 regionshigh availability$0.74/hrlaunch
L40Scommunity48G4 regionsavailable$0.79/hrlaunch
H100 PCIE80G2 regionsavailable$1.98/hrlaunch
RTX A4000community16G5 regionsavailable$0.08/hrlaunch
RTX 5080community16G7 regionsavailable$0.16/hrlaunch
A100 40GBcommunity40G6 regionsavailable$0.56/hrlaunch
RTX 509032G2 regionsavailable$0.99/hrlaunch
L424G4 regionsavailable$0.44/hrlaunch
RTX 3080community10G5 regionsavailable$0.10/hrlaunch
H200 NVLcommunity141G3 regionsavailable$3.98/hrlaunch
RTX 6000 ADAcommunity48G2 regionsavailable$0.74/hrlaunch
B300288G2 regionsavailable$7.89/hrlaunch
H100 PCIEcommunity80G2 regionsavailable$2.54/hrlaunch
H200 SXMcommunity141G4 regionsavailable$3.98/hrlaunch
H100 SXMcommunity80G3 regionsrunning low$2.27/hrlaunch
A4048G2 regionsrunning low$0.49/hrlaunch
RTX 309024G2 regionsrunning low$0.50/hrlaunch
B200192G2 regionsrunning low$6.79/hrlaunch
RTX A5000community24G3 regionsrunning low$0.23/hrlaunch
L4community24G2 regionsrunning low$0.27/hrlaunch
B200community192G2 regionsrunning low$5.32/hrlaunch
RTX A6000community48G2 regionsrunning low$0.41/hrlaunch
RTX 4080community16G3 regionsrunning low$0.18/hrlaunch
RTX PRO 6000community96G1 regionrunning low$1.69/hrlaunch
A100 40GB40G1 regionrunning low$0.89/hrlaunch
V100community16G2 regionsrunning low$0.09/hrlaunch

lowest live on-demand rate per GPU, refreshed continuously · per-second billing

H100 NVL price comparison (September 2026)

ProviderPrice / GPU / hrNotes
GPU.ai (Secure Cloud)$3.19/hrVetted data center capacity, per-second billing
GPU.ai (Community Cloud)$2.37/hrMarketplace capacity from independent suppliers
CoreWeavePublic on-demand list rate
AWSPublic on-demand list rate
AzurePublic on-demand list rate

competitor rates are public on-demand list prices, refreshed daily · gpu.ai prices are live floor rates with zero markup

Compare and explore

faq

H100 NVL rental questions

How much does it cost to rent an H100 NVL in the cloud?

H100 NVL cloud instances on GPU.ai currently start at $2.37 per GPU per hour, billed per second. That is the live floor price across every provider in our fleet. We route your launch to the cheapest provider with stock and add zero markup.

How much VRAM does the H100 NVL have?

The NVIDIA H100 NVL has 94 GB of HBM3 memory with up to 3,900 GB/s of bandwidth. GPU.ai reports the minimum VRAM actually delivered across our fleet, so the figure you see at launch is the figure you get.

How does GPU.ai's H100 NVL pricing compare to AWS, Azure, and CoreWeave?

GPU.ai aggregates a dozen-plus GPU clouds and passes through each provider's raw price with no markup, so the same silicon is typically far cheaper than hyperscaler list rates. The comparison table on this page shows the current public on-demand list prices side by side.

How does billing work?

Billing is per second, starting when your instance is SSH-ready and stopping the moment you terminate. There are no minimum commitments, no reservation fees, and no egress surprises: you pay the listed hourly rate pro-rated to the second.