| Precision | Peak (TFLOPS) | Per GPU (TFLOPS) | TFLOPS/W |
|---|---|---|---|
| FP64 | 240 | 30 | — |
| FP32 | 480 | 60 | — |
| TF32 | 7,000 | 875 | — |
| FP16 | 14,000 | 1,750 | — |
| BF16 | 14,000 | 1,750 | — |
| FP8 | 28,000 | 3,500 | — |
| FP4 | 56,000 | 7,000 | — |
| INT8 | 28,000 | 3,500 | — |
All figures are dense. Vendors commonly headline the sparse number, which is twice the dense one.
NVSwitch + NVLink