Pricing
Transparent, per-second pricing on the best-available GPU across every provider we aggregate. No minimums, no lock-in, no surprises.
Real prices. No hidden fees. See how GPU.ai stacks up.
| GPU Model | VRAM | GPU.ai | CoreWeave | AWS | Azure | Avg Savings |
|---|---|---|---|---|---|---|
| H200 SXM | 141 GB | $3.81/hr | $6.31/hr | $4.97/hr | $12.99/hr | 53% |
| H100 SXM | 80 GB | $2.64/hr | $6.16/hr | $3.93/hr | $12.29/hr | 65% |
| B200 | 192 GB | $5.29/hr | $8.60/hr | $14.24/hr | $15.00/hr | 58% |
| A100 | 80 GB | $2.00/hr | $2.70/hr | $5.12/hr | $4.10/hr | 48% |
Starting rates across the rest of the fleet. Live rates and stock update continuously.
| Model | VRAM | GPU.ai / hr | Avg savings |
|---|---|---|---|
| H100 PCIe | 80 GB HBM3 | $2.49/hr | 60% |
| A100 40GB | 40 GB HBM2e | $1.45/hr | 45% |
| L40S | 48 GB GDDR6 | $1.10/hr | 40% |
| RTX 6000 Ada | 48 GB GDDR6 | $0.95/hr | 38% |
| A6000 | 48 GB GDDR6 | $0.79/hr | 32% |
| RTX 4090 | 24 GB GDDR6X | $0.44/hr | 35% |
| L4 | 24 GB GDDR6 | $0.42/hr | 33% |
| V100 | 16 GB HBM2 | $0.29/hr | 25% |
Starting rates shown. See live availability →
Metering starts when your instance hits SSH-ready and stops the second you do. No hourly rounding, no idle reservation fees.
Every launch is matched to the cheapest provider that has stock for your exact spec — automatically, on every request.
Move between providers and regions whenever you like. The CLI and API stay identical, so your workflow never changes.
New accounts get $100 in compute credit. No credit card required, and nothing auto-charges when it runs out.
Per second, from the moment your instance reaches SSH-ready to the moment you stop it. No hourly rounding, no minimums, no idle reservation fees.
We continuously scan live capacity across every connected provider and route each launch to the cheapest one that matches your spec — so you get the floor price without checking five dashboards.
No surprise networking fees. You pay the GPU rate; data transfer for normal training and inference workloads is included.
$100 of compute credit applied to any on-demand GPU. No credit card required to start, and nothing auto-charges when the credit runs out.
Yes. Reserved instances trade a commitment for a deeper discount on multi-week and multi-month workloads — see the Reserved product.