The same infrastructure, for less than the leading cloud charges.
Every figure below is the leading cloud's own published on-demand price, beside ours. Like for like: the same GPU, its cheapest machine with the same vCPUs and memory, the same class of service.
What a month would cost
Change the quantities to match your own workload.
CPU instances
| Compute | Leading cloud | You save |
|---|---|---|
| $0.0235/hr · cx22 | $0.0336/hr Its cheapest machine with 2 vCPU and 4 GiB | 30% |
| $0.0941/hr · cx32 | $0.1344/hr Its cheapest machine with 4 vCPU and 16 GiB | 30% |
| $0.0941/hr · cpx31 | $0.1344/hr Its cheapest machine with 4 vCPU and 16 GiB | 30% |
| $0.1882/hr · cx52 | $0.2688/hr Its cheapest machine with 8 vCPU and 32 GiB | 30% |
| $0.1882/hr · cpx51 | $0.2688/hr Its cheapest machine with 8 vCPU and 32 GiB | 30% |
GPU instances
| Compute | Leading cloud | You save |
|---|---|---|
| $1.92/hr · A100 40GB | $2.74/hr Same GPU (A100 40 GB) | 30% |
| $2.40/hr · A100 80GB | $3.43/hr Same GPU (A100 80 GB) | 30% |
| $2.40/hr · A100 PCIe | $3.43/hr Same GPU (A100 80 GB) | 30% |
| $0.5634/hr · RTX 3090 | $0.8048/hr Its nearest GPU (L4) — it does not sell this one | 30% |
| $1.30/hr · L40S | $1.86/hr Same GPU (L40S) | 30% |
| $4.82/hr · H100 SXM | $6.88/hr Same GPU (H100 80 GB) | 30% |
| $4.82/hr · H100 PCIe | $6.88/hr Same GPU (H100 80 GB) | 30% |
| $0.7646/hr · RTX 4090 | $0.8048/hr Its nearest GPU (L4) — it does not sell this one | 5% |
CosmoBase
| Compute | Leading cloud | You save |
|---|---|---|
| $0.0112 / database-hour · Database | $0.016 / database-hour Managed database, smallest instance | 30% |
Cloud Services
| Compute | Leading cloud | You save |
|---|---|---|
| $0.0218 / GB-month · Object storage | $0.023 / GB-month Standard object storage — Compared with its first-tier price, which it lowers at higher volumes. | 5% |
Panel
| Compute | Leading cloud | You save |
|---|---|---|
| $0.007 / build minute · Build minutes | $0.01 / build minute Managed hosting build minutes | 30% |
AI inference
| Compute | Leading cloud | You save |
|---|---|---|
| $0.455 / million tokens · General & code model — output | $0.65 / million tokens The same open-weight model, output tokens | 30% |
| $0.105 / million tokens · General & code model — input | $0.15 / million tokens The same open-weight model, input tokens | 30% |
Support
| Compute | Leading cloud | You save |
|---|---|---|
| $70.00 / month · Business support | $100 / month Business support plan, monthly minimum | 30% |
| $3,850 / month · Enterprise support | $5,500 / month Enterprise support plan, monthly minimum | 30% |
| $20.30 / month · Developer support | $29.00 / month Developer support plan, monthly minimum | 30% |
“The leading cloud” is the largest public cloud provider by market share. Its prices here are its published on-demand list prices for its US East region, Linux, as published on 10 November 2026; machine prices are re-read from its price list daily. Where it tiers a price by volume, the comparison is with its first tier. Commitment discounts, reserved capacity and spot pricing are not included on either side.
Not every Compute price is lower: only prices that are appear here. The price list has every price.