A10G vs RTX 5090
The cheapest RTX 5090 ($0.22) costs 78% less per hour than the cheapest A10G ($1.00).
Per unit of performance, the RTX 5090 is 94% cheaper: $0.24 versus $4.02 per A100-equivalent hour.
The RTX 5090 has 32 GB of memory against 24 GB on the A10G.
| Measure | NVIDIA A10G 24GB (AWS) | NVIDIA GeForce RTX 5090 |
|---|---|---|
| Lowest per GPU-hour | $1.00 | $0.22 |
| Median per GPU-hour | $1.44 | $0.87 |
| On-demand from | $1.00 | $0.34 |
| Spot from | — | $0.25 |
| Reserved from | — | $0.22 |
| Lowest per A100-equivalent hour | $4.02 | $0.24 |
| Live offers | 22 | 405 |
| Providers | 3 | 84 |
| Memory | 24 GB GDDR6 | 32 GB GDDR7 |
| Memory bandwidth | 600 GB/s | 1,792 GB/s |
| FP16 / BF16 | 70 TFLOPS | 209.5 TFLOPS |
| FP8 | — | 419 TFLOPS |
| Performance vs A100 80GB | 0.25× | 0.92× |
| Architecture | Ampere | Blackwell |
| Power | 150 W | 575 W |