A10 vs RTX 5090
The cheapest RTX 5090 ($0.22) costs 3% less per hour than the cheapest A10 ($0.22).
Per unit of performance, the RTX 5090 is 63% cheaper: $0.24 versus $0.63 per A100-equivalent hour.
The RTX 5090 has 32 GB of memory against 24 GB on the A10.
| Measure | NVIDIA A10 24GB | NVIDIA GeForce RTX 5090 |
|---|---|---|
| Lowest per GPU-hour | $0.22 | $0.22 |
| Median per GPU-hour | $1.41 | $0.87 |
| On-demand from | $0.24 | $0.34 |
| Spot from | $0.59 | $0.25 |
| Reserved from | $0.22 | $0.22 |
| Lowest per A100-equivalent hour | $0.63 | $0.24 |
| Live offers | 302 | 405 |
| Providers | 25 | 84 |
| Memory | 24 GB GDDR6 | 32 GB GDDR7 |
| Memory bandwidth | 600 GB/s | 1,792 GB/s |
| FP16 / BF16 | 125 TFLOPS | 209.5 TFLOPS |
| FP8 | — | 419 TFLOPS |
| Performance vs A100 80GB | 0.35× | 0.92× |
| Architecture | Ampere | Blackwell |
| Power | 150 W | 575 W |