Hyperstack vs Spheron
Hyperstack is cheaper on 4 of the 5 GPUs both providers list with on-demand pricing, verified 2026-07-18. Spheron still wins on one, its best case being the H100 at $2.01 vs $2.50. Platform facts come from each provider's own documentation.
Price per GPU, side by side (2026-07-18)
| GPU (on-demand $/hr) | Hyperstack | Spheron |
|---|---|---|
| H100 | $2.50 | $2.01 |
| A100 | $1.35 | $1.43 |
| H200 | $3.99 | $4.54 |
| B200 | $6.00 | $9.36 |
| L40S | not offered | $0.96 |
| B300 | $7.40 | not offered |
| L40 | $1.00 | not offered |
| GH200 | not offered | $1.88 |
| RTX PRO 6000 | $1.85 | $2.39 |
| RTX A6000 | $0.50 | not offered |
| RTX A4000 | $0.15 | not offered |
| RTX 4090 | not offered | $0.53 |
| RTX 5090 | not offered | $0.86 |
Cheapest verified on-demand rate per side from today's snapshot; highlighted = cheaper. Spot/marketplace tiers shown on the individual pages.
Platform facts
| Hyperstack | Spheron | |
|---|---|---|
| Billing increment | not published | not published |
| Egress | not published | not published |
| Storage /GB-mo | $0.07/GB/month shared storage | not published |
| Spot offered | yes | yes |
| Reserved offered | not published | no |
| Notebook support | templates | templates |
From each provider's documentation, verified 2026-07-07; sources on the fees matrix and enterprise matrix.
FAQ
Hyperstack is cheaper on 4 of the 5 GPUs both providers offer, verified 2026-07-18. The other is cheaper on the rest, so the right pick depends on the specific GPU you need.
The B200: $6.00 per GPU-hour on Hyperstack vs $9.36 on Spheron, a difference of $3.36 per hour, verified 2026-07-18.
Hyperstack lists 4 GPUs with on-demand pricing that Spheron does not: B300, L40, RTX A6000, RTX A4000. Spheron lists 4 that Hyperstack does not: L40S, GH200, RTX 4090, RTX 5090.
Prices render from the live dataset each build. More: the screener · price movers · methodology.