H100 SXM 80GB vs RTX 5000 ADA 32GB
The cheapest RTX 5000 ADA 32GB ($0.42) costs 16% less per hour than the cheapest H100 SXM 80GB ($0.50).
Per unit of performance, the H100 SXM 80GB is 74% cheaper: $0.17 versus $0.63 per A100-equivalent hour.
The H100 SXM 80GB has 80 GB of memory against 32 GB on the RTX 5000 ADA 32GB.
| Measure | NVIDIA H100 80GB SXM5 | NVIDIA RTX 5000 Ada Generation 32GB |
|---|---|---|
| Lowest per GPU-hour | $0.50 | $0.42 |
| Median per GPU-hour | $3.51 | $0.84 |
| On-demand from | $0.58 | $0.42 |
| Spot from | $0.50 | — |
| Reserved from | $1.41 | $0.56 |
| Lowest per A100-equivalent hour | $0.17 | $0.63 |
| Live offers | 317 | 18 |
| Providers | 83 | 10 |
| Memory | 80 GB HBM3 | 32 GB GDDR6 |
| Memory bandwidth | 3,350 GB/s | 576 GB/s |
| FP16 / BF16 | 989 TFLOPS | 261.1 TFLOPS |
| FP8 | 1,979 TFLOPS | 522.2 TFLOPS |
| Performance vs A100 80GB | 3× | 0.67× |
| Architecture | Hopper | Ada Lovelace |
| Power | 700 W | 250 W |