P4 vs RTX 5090
The cheapest P4 ($0.01) costs 95% less per hour than the cheapest RTX 5090 ($0.22).
Per unit of performance, the RTX 5090 is 25% cheaper: $0.24 versus $0.32 per A100-equivalent hour.
The RTX 5090 has 32 GB of memory against 8 GB on the P4.
| Measure | NVIDIA Tesla P4 8GB | NVIDIA GeForce RTX 5090 |
|---|---|---|
| Lowest per GPU-hour | $0.01 | $0.22 |
| Median per GPU-hour | $0.86 | $0.87 |
| On-demand from | $0.01 | $0.34 |
| Spot from | — | $0.25 |
| Reserved from | $0.34 | $0.22 |
| Lowest per A100-equivalent hour | $0.32 | $0.24 |
| Live offers | 46 | 405 |
| Providers | 7 | 84 |
| Memory | 8 GB GDDR5 | 32 GB GDDR7 |
| Memory bandwidth | 192 GB/s | 1,792 GB/s |
| FP16 / BF16 | — | 209.5 TFLOPS |
| FP8 | — | 419 TFLOPS |
| Performance vs A100 80GB | 0.03× | 0.92× |
| Architecture | Pascal | Blackwell |
| Power | 75 W | 575 W |