Independent reference. We are independent of every vendor listed. No affiliate links. No sponsored placements.
H100 cost per hour - Last verified June 2026

Nvidia H100 cost per hour across 17 GPU cloud vendors

Per-GPU per-hour H100 rates pulled from the public pricing pages on the 17 vendor pages on this site. Sorted cheapest first. Quote-only vendors are listed at the bottom and we do not infer rates for them.

VendorPricing modelConfigurationPer GPU per hour
Vast.aimarketplaceMarketplace floor, interruptible$1.490
DigitalOcean GPU Dropletson-demand12-month reserved, per-GPU on 8x$1.990
Hyperstackon-demandOn-demand$2.400
RunPodon-demandSecure Cloud, per-GPU$2.990
Lambda Labson-demandOn-demand, per-GPU on 8x$3.290
Together AIon-demandReserved cluster, per-GPU$3.290
Modalper-secondPer-second, hourly equiv$3.950
Crusoe Cloudon-demandOn-demand, per-GPU on 8x$4.650
Replicateper-secondPer-second, hourly equiv$5.040
Paperspaceon-demandPer-GPU on 8x$5.950
CoreWeaveon-demandHGX H100 8x node, per-GPU$6.155
Oracle Cloud (OCI)on-demandBare metal, per-GPU equiv$10.000
Google Cloudon-demandOn-demand, per-GPU equiv$11.060
AWS EC2on-demandOn-demand, 8x per node, per-GPU equiv$12.288
Azureon-demandOn-demand, per-GPU equiv$12.290
Fly.io GPUquote-onlyQuote only - contact Fly.io GPU salesQuote only
OVHcloud GPUquote-onlyQuote only - contact OVHcloud GPU salesQuote only

How to read this table

Per-GPU per-hour means the rate divided across all GPUs in the smallest billable unit (typically an 8-GPU HGX node). Hyperscaler on-demand list rates are the published rate before any negotiated discount, Savings Plan, Reserved Instance, or Capacity Blocks commitment. Specialist clouds (CoreWeave, Lambda, Together AI) usually publish reservation rates that are 30 to 60 percent below their on-demand. Marketplace floor (Vast.ai) is interruptible.

Last verified June 2026.