A30 vs L40S
The cheapest A30 ($0.15) costs 61% less per hour than the cheapest L40S ($0.38).
Per unit of performance, the A30 is 24% cheaper: $0.30 versus $0.40 per A100-equivalent hour.
The L40S has 48 GB of memory against 24 GB on the A30.
| Measure | NVIDIA A30 24GB | NVIDIA L40S 48GB |
|---|---|---|
| Lowest per GPU-hour | $0.15 | $0.38 |
| Median per GPU-hour | $0.64 | $1.30 |
| On-demand from | $0.22 | $0.38 |
| Spot from | $0.15 | $0.45 |
| Reserved from | $0.23 | $0.49 |
| Lowest per A100-equivalent hour | $0.30 | $0.40 |
| Live offers | 32 | 284 |
| Providers | 17 | 85 |
| Memory | 24 GB HBM2 | 48 GB GDDR6 |
| Memory bandwidth | 933 GB/s | 864 GB/s |
| FP16 / BF16 | 165 TFLOPS | 362 TFLOPS |
| FP8 | — | 733 TFLOPS |
| Performance vs A100 80GB | 0.5× | 0.96× |
| Architecture | Ampere | Ada Lovelace |
| Power | 165 W | 350 W |