P100 12GB vs RTX 4090
The cheapest P100 12GB ($0.069) costs 57% less per hour than the cheapest RTX 4090 ($0.16).
Per unit of performance, the RTX 4090 is 58% cheaper: $0.25 versus $0.60 per A100-equivalent hour.
The RTX 4090 has 24 GB of memory against 12 GB on the P100 12GB.
| Measure | NVIDIA Tesla P100 12GB PCIe | NVIDIA GeForce RTX 4090 |
|---|---|---|
| Lowest per GPU-hour | $0.069 | $0.16 |
| Median per GPU-hour | $0.08 | $0.58 |
| On-demand from | $0.069 | $0.19 |
| Spot from | — | $0.16 |
| Reserved from | — | $0.18 |
| Lowest per A100-equivalent hour | $0.60 | $0.25 |
| Live offers | 5 | 427 |
| Providers | 4 | 91 |
| Memory | 12 GB HBM2 | 24 GB GDDR6X |
| Memory bandwidth | 549 GB/s | 1,008 GB/s |
| FP16 / BF16 | 21.2 TFLOPS | 165.2 TFLOPS |
| FP8 | — | 330.3 TFLOPS |
| Performance vs A100 80GB | 0.12× (estimate) | 0.63× |
| Architecture | Pascal | Ada Lovelace |
| Power | 250 W | 450 W |