GPU EconomyCloud GPU prices, read every hour
Latest read 110 providers tracked18,940 prices read in 24 h

Renting an H100 SXM costs from $1.79 an hour today.

NVIDIA H100 SXM, 80 GB · on-demand at Cudo Compute

On-demand · low risk — 41 rates, cheapest firstUSD per GPU-hour
ProviderRateGo
Cudo Compute$1.79Deploy
Seeweb$2.11+$0.32 vs cheapestDeploy
Bitdeer AI$2.36+$0.57 vs cheapestDeploy
HPC-AI.COM$2.42+$0.63 vs cheapestDeploy
Denvr Dataworks$2.45+$0.66 vs cheapestDeploy
Taiga Cloud (Northern Data Group)$2.60+$0.81 vs cheapestDeploy
Atlas Cloud$2.95+$1.16 vs cheapestDeploy
Limestone Networks$2.95+$1.16 vs cheapestDeploy
VESSL AI$2.98+$1.19 vs cheapestDeploy
Civo$2.99+$1.20 vs cheapestDeploy
FPT AI Factory$2.99+$1.20 vs cheapestDeploy
HexGrid$2.99+$1.20 vs cheapestDeploy
Geodd$3.00+$1.21 vs cheapestDeploy
Massed Compute$3.14+$1.35 vs cheapestDeploy
Hyperbolic$3.19+$1.40 vs cheapestDeploy
Hyperstack$3.20+$1.41 vs cheapestDeploy
Thunder Compute$3.20+$1.41 vs cheapestDeploy
Canopy Wave$3.25+$1.46 vs cheapestDeploy
FarmGPU$3.29+$1.50 vs cheapestDeploy
Jarvislabs$3.49+$1.70 vs cheapestDeploy
Scaleway$3.54+$1.75 vs cheapestDeploy
Elice Cloud$3.57+$1.78 vs cheapestDeploy
Barrack AI$3.59+$1.80 vs cheapestDeploy
Crusoe$3.90+$2.11 vs cheapestDeploy
GPU.ai$3.99+$2.20 vs cheapestDeploy
Lambda$3.99+$2.20 vs cheapestDeploy
RunPod$3.99+$2.20 vs cheapestDeploy
Together AI$3.99+$2.20 vs cheapestDeploy
Neysa$4.39+$2.60 vs cheapestDeploy
Paperspace$4.41+$2.62 vs cheapestDeploy
Nebius$4.50+$2.71 vs cheapestDeploy
Verda$4.61+$2.82 vs cheapestDeploy
Tata Communications$5.13+$3.34 vs cheapestDeploy
FluidCore$5.52+$3.73 vs cheapestDeploy
Paperspace$5.95+$4.16 vs cheapestDeploy
CoreWeave$6.16+$4.37 vs cheapestDeploy
Amazon Web Services$6.88+$5.09 vs cheapestDeploy
STACKIT$9.00+$7.21 vs cheapestDeploy
Google Cloud$9.80+$8.01 vs cheapestDeploy
Oracle Cloud (OCI)$10.00+$8.21 vs cheapestDeploy
Microsoft Azure$11.06+$9.27 vs cheapestDeploy
Spot · high risk — 6 rates, cheapest firstUSD per GPU-hour
ProviderRateGo
Google Cloud$1.06Deploy
Microsoft Azure$2.04−$9.02 (−82%) vs its on-demand $11.06Deploy
Verda$2.30−$2.30 (−50%) vs its on-demand $4.61Deploy
CoreWeave$2.44−$3.72 (−60%) vs its on-demand $6.16Deploy
CoreWeave$2.46−$3.69 (−60%) vs its on-demand $6.16Deploy
Elice Cloud$3.21−$0.36 (−10%) vs its on-demand $3.57Deploy
Committed · low risk — 12 rates, cheapest firstUSD per GPU-hour
ProviderRateGo
Bitdeer AI$1.79min. commitment 3yr · −$0.57 (−24%) vs its on-demand $2.36Reserve
Seeweb$1.80min. commitment 1yr · −$0.31 (−15%) vs its on-demand $2.11Reserve
Bitdeer AI$2.08min. commitment 1yr · −$0.28 (−12%) vs its on-demand $2.36Reserve
Civo$2.49min. commitment 3yr · −$0.50 (−17%) vs its on-demand $2.99Reserve
Civo$2.69min. commitment 1yr · −$0.30 (−10%) vs its on-demand $2.99Reserve
Hyperstack$2.72min. commitment commit · +$0.93 vs cheapestReserve
Together AI$3.19min. commitment commit · +$1.40 vs cheapestReserve
Paperspace$3.26min. commitment 12mo · −$1.15 (−26%) vs its on-demand $4.41Reserve
Google Cloud$4.30min. commitment 3yr · +$2.51 vs cheapestReserve
Microsoft Azure$4.86min. commitment 3yr · −$6.21 (−56%) vs its on-demand $11.06Reserve
Google Cloud$6.80min. commitment 1yr · +$5.01 vs cheapestReserve
Microsoft Azure$7.08min. commitment 1yr · −$3.98 (−36%) vs its on-demand $11.06Reserve
Other · high to medium risk — 3 rates, cheapest firstUSD per GPU-hour
ProviderRateGo
RunPod$2.69communityDeploy
Lium$3.75community · +$1.06 vs cheapestDeploy
Modal$3.95serverless · +$1.26 vs cheapestDeploy

Your job

The list re-ranks by the total, whole nodes counted where a provider rents only nodes.

GPUs
How long
Market

Choose a card

8 GPUs for a month, 24/7 (730 h) cost from $10,454 at the listed on-demand rate of Cudo Compute ($1.79 per GPU-hour); the job bar above ranks every provider for a job of your own.

In the last closed month (September 2026) the H100 SXM like-for-like median rose 2.6%, the second month running. Over 12 months, +19%.

Add your GPU cloudPrices come from your public pricing page, never from this form. Listing is free and can’t be bought. How providers are listed, and who is not, and why
Your email is used only about your listing, never shared.

Data updated: