Google Cloud vs Hyperstack
Hyperstack is cheaper on 4 of the 5 GPU configurations both providers list with on-demand pricing, updated 2026-10-02. Google Cloud still wins on one, its best case being the RTX PRO 6000 (unspecified) · node count not reported at $1.10 vs $1.85. Platform facts come from each provider's documentation.
Google Cloud
cheaper on 1 of 5 sharedHyperstack
cheaper on 4 of 5 sharedPrices are per GPU-hour. Rows match exact GPU variants and published node minimums. When node count is not reported, confirm the required configuration before comparing whole-job costs. These are listed rates, not a guarantee of stock.
Price per GPU, side by side (2026-10-02)
| GPU (on-demand $/hr) | Google Cloud | Hyperstack | Price diff |
|---|---|---|---|
| RTX A4000 · node count not reported | not offered | $0.15 | – |
| Tesla T4 · node count not reported | $0.35 | not offered | – |
| RTX A6000 · node count not reported | not offered | $0.50 | – |
| L4 · node count not reported | $0.560 | not offered | – |
| L40 · node count not reported | not offered | $1.00 | – |
| RTX PRO 6000 (unspecified) · node count not reported | $1.10 | $1.85 | -$0.754 (-40.8%) Google Cloud |
| A100 (unspecified) · node count not reported | not offered | $1.35 | – |
| A100 SXM 80GB · node count not reported | $3.93 | $1.60 | -$2.33 (-59.3%) Hyperstack |
| Tesla V100 (unspecified) · node count not reported | $2.48 | not offered | – |
| H100 (unspecified) · node count not reported | $9.80 | $2.50 | -$7.30 (-74.5%) Hyperstack |
| A100 SXM 40GB · node count not reported | $2.93 | not offered | – |
| H100 SXM · node count not reported | not offered | $3.20 | – |
| H200 (SXM) · node count not reported | $9.31 | $3.99 | -$5.32 (-57.2%) Hyperstack |
| B200 · node count not reported | $16.11 | $6.00 | -$10.11 (-62.8%) Hyperstack |
| B300 · node count not reported | not offered | $7.40 | – |
Lowest recorded on-demand rate per configuration; highlighted = cheaper. Price diff is the absolute and percent saving on the cheaper side (named), relative to the higher price; – means one side has no matching configuration in the collected rates. Spot/marketplace tiers shown on the individual pages.
Tables rank by price, lowest first. Commercial relationships never affect ordering.How we collect and rank prices.
Platform facts
| Google Cloud | Hyperstack | |
|---|---|---|
| Billing increment | not recorded | not published |
| Egress | not recorded | not published |
| Storage /GB-mo | not recorded | $0.07/GB/month shared storage |
| Spot offered | not recorded | yes |
| Reserved offered | not recorded | not published |
| Notebook support | not recorded | templates |
From each provider's documentation, updated 2026-07-07; sources on the fees matrix and enterprise matrix.
FAQ
Hyperstack is cheaper on 4 of the 5 GPU configurations both providers list, updated 2026-10-02. Google Cloud is cheaper on 1; 0 have equal rates. Compare the configuration you need.
The B200 · node count not reported: $6.00 per GPU-hour on Hyperstack vs $16.11 on Google Cloud, a difference of $10.11 per hour, updated 2026-10-02.
Google Cloud lists 4 GPU configurations with on-demand pricing that Hyperstack does not: Tesla T4 · node count not reported, L4 · node count not reported, Tesla V100 (unspecified) · node count not reported, A100 SXM 40GB · node count not reported. Hyperstack lists 6 that Google Cloud does not: RTX A4000 · node count not reported, RTX A6000 · node count not reported, L40 · node count not reported, A100 (unspecified) · node count not reported, H100 SXM · node count not reported, B300 · node count not reported.
Compare more rental options: the screener · price movers · methodology.