A30 vs RTX 5090
The cheapest A30 ($0.15) costs 31% less per hour than the cheapest RTX 5090 ($0.22).
Per unit of performance, the RTX 5090 is 22% cheaper: $0.24 versus $0.30 per A100-equivalent hour.
The RTX 5090 has 32 GB of memory against 24 GB on the A30.
| Measure | NVIDIA A30 24GB | NVIDIA GeForce RTX 5090 |
|---|---|---|
| Lowest per GPU-hour | $0.15 | $0.22 |
| Median per GPU-hour | $0.64 | $0.87 |
| On-demand from | $0.22 | $0.34 |
| Spot from | $0.15 | $0.25 |
| Reserved from | $0.23 | $0.22 |
| Lowest per A100-equivalent hour | $0.30 | $0.24 |
| Live offers | 32 | 405 |
| Providers | 17 | 84 |
| Memory | 24 GB HBM2 | 32 GB GDDR7 |
| Memory bandwidth | 933 GB/s | 1,792 GB/s |
| FP16 / BF16 | 165 TFLOPS | 209.5 TFLOPS |
| FP8 | — | 419 TFLOPS |
| Performance vs A100 80GB | 0.5× | 0.92× |
| Architecture | Ampere | Blackwell |
| Power | 165 W | 575 W |