H100 NVL vs V100: Specs
Architecture, compute capability and supported precisions for the H100 NVL and the V100, from vendor datasheets.
No live price to compare
No provider on Aquanode is listing at least one of these two cards right now, so no hourly rate is shown. This page shows live prices as soon as both are listed.
Specs side by side
| Spec | H100 NVL | V100 |
|---|---|---|
| VRAM | 94GB HBM3 | 16GB or 32GB HBM2 |
| Memory bandwidth | 3.9 TB/s | 900 GB/s |
| Architecture | Hopper | Volta |
| Compute capability | 9.0 | 7.0 |
| Precisions in hardware | FP32, FP16, BF16, FP8, INT4 | FP32, FP16 |
| Peak FP16/BF16 tensor TFLOPS | 835.5 TFLOPS | 125 TFLOPS (SXM2), 112 TFLOPS (PCIe) |
| Peak FP8 tensor TFLOPS | 1,670.5 TFLOPS | – |
| NVLink / multi-GPU interconnect | NVLink, 600 GB/s | NVLink, 300 GB/s (SXM2 only) |
| Max board power (TDP) | 350-400W (configurable) | 300W (SXM2), 250W (PCIe) |
