Benchmark results across SScoreCompute's full GPU lineup — tested on real AI workloads. All prices in CAD.
| SPECIFICATION | H100 | H200 | B300 | GB300 NVL72 |
|---|---|---|---|---|
| Architecture | Hopper | Hopper | Blackwell | Blackwell Ultra |
| VRAM | 80GB HBM3 | 141GB HBM3e | 192GB HBM3e | 13.8TB pooled |
| Memory Bandwidth | 3.35 TB/s | 4.8 TB/s | 8.0 TB/s | 130 TB/s NVLink |
| FP8 Performance | 3,958 TFLOPS | 3,958 TFLOPS | 9,000 TFLOPS | 648,000 TFLOPS |
| NVLink | NVLink 4.0 | NVLink 4.0 | NVLink 5.0 | NVLink 5.0 · 72-GPU |
| Power (TDP) | 700W | 700W | 1,000W | 120kW rack |
| Price (CAD/hr) | $3.14 | $4.10 | $6.71 | $409 |
| Price (USD/hr) | ~$2.29 | ~$2.99 | ~$4.90 | ~$299 |
| Availability | On-demand | On-demand | On-demand | Reserve |
All benchmarks run on SScoreCompute infrastructure. Results may vary by model configuration and batch size.