P4 vs RTX 4090
The cheapest P4 ($0.01) costs 93% less per hour than the cheapest RTX 4090 ($0.16).
Per unit of performance, the RTX 4090 is 20% cheaper: $0.25 versus $0.32 per A100-equivalent hour.
The RTX 4090 has 24 GB of memory against 8 GB on the P4.
| Measure | NVIDIA Tesla P4 8GB | NVIDIA GeForce RTX 4090 |
|---|---|---|
| Lowest per GPU-hour | $0.01 | $0.16 |
| Median per GPU-hour | $0.86 | $0.58 |
| On-demand from | $0.01 | $0.19 |
| Spot from | — | $0.16 |
| Reserved from | $0.34 | $0.18 |
| Lowest per A100-equivalent hour | $0.32 | $0.25 |
| Live offers | 46 | 427 |
| Providers | 7 | 91 |
| Memory | 8 GB GDDR5 | 24 GB GDDR6X |
| Memory bandwidth | 192 GB/s | 1,008 GB/s |
| FP16 / BF16 | — | 165.2 TFLOPS |
| FP8 | — | 330.3 TFLOPS |
| Performance vs A100 80GB | 0.03× | 0.63× |
| Architecture | Pascal | Ada Lovelace |
| Power | 75 W | 450 W |