NVIDIA H100 NVL vs NVIDIA Rubin CPX
How the NVIDIA H100 NVL and NVIDIA Rubin CPX compare across memory, compute, power and form factor, with the stronger figure highlighted on each row.
| Specification |
HOPPER
NVIDIA H100 NVL
View Product →
|
RUBIN
NVIDIA Rubin CPX
View Product →
|
|---|---|---|
| Pricing | ||
| Price | $39,600.00 | – |
| Overview | ||
| Architecture | Hopper | Rubin (CPX variant) |
| Memory | ||
| Memory | 94 GB HBM3 | HBM4 (capacity per partner configuration) |
| Memory Bandwidth | 3.9 TB/s | – |
| Compute | ||
| FP32 | 60 TFLOPS | – |
| FP64 Tensor | 30 TFLOPS / 60 TFLOPS | – |
| TF32 Tensor | 835 TFLOPS (with sparsity) | – |
| FP16/BF16 Tensor | 1,671 TFLOPS (with sparsity) | – |
| FP8 Tensor | 3,341 TFLOPS (with sparsity) | – |
| INT8 Tensor | 3,341 TOPS | – |
| Platform | ||
| NVLink | NVLink bridges, PCIe Gen5 | NVLink 6, PCIe Gen 6 |
| TDP | 350–400W (configurable) | – |
| Form Factor | Dual PCIe dual-slot air cooled | SXM / rack-integrated |
Not sure which fits?
Talk to an Expert →
Talk it through with a specialist
Tell us your workload and we will recommend the right GPU, configuration, and sourcing options.