RTX 5090 vs V100 16GB
The cheapest V100 16GB ($0.052) costs 76% less per hour than the cheapest RTX 5090 ($0.22).
Per unit of performance, the V100 16GB is 47% cheaper: $0.12 versus $0.24 per A100-equivalent hour.
The RTX 5090 has 32 GB of memory against 16 GB on the V100 16GB.
| Measure | NVIDIA GeForce RTX 5090 | NVIDIA Tesla V100 16GB |
|---|---|---|
| Lowest per GPU-hour | $0.22 | $0.052 |
| Median per GPU-hour | $0.87 | $0.77 |
| On-demand from | $0.34 | $0.052 |
| Spot from | $0.25 | $0.13 |
| Reserved from | $0.22 | $0.16 |
| Lowest per A100-equivalent hour | $0.24 | $0.12 |
| Live offers | 405 | 196 |
| Providers | 84 | 40 |
| Memory | 32 GB GDDR7 | 16 GB HBM2 |
| Memory bandwidth | 1,792 GB/s | 900 GB/s |
| FP16 / BF16 | 209.5 TFLOPS | 125 TFLOPS |
| FP8 | 419 TFLOPS | — |
| Performance vs A100 80GB | 0.92× | 0.42× |
| Architecture | Blackwell | Volta |
| Power | 575 W | 300 W |