NVIDIA H100 SXM vs NVIDIA Rubin CPX
How the NVIDIA H100 SXM and NVIDIA Rubin CPX compare across memory, compute, power and form factor, with the stronger figure highlighted on each row.
| Specification |
HOPPER
NVIDIA H100 SXM
View Product →
|
RUBIN
NVIDIA Rubin CPX
View Product →
|
|---|---|---|
| Pricing | ||
| Price | – | – |
| Overview | ||
| Architecture | Hopper | Rubin (CPX variant) |
| Memory | ||
| Memory | 80 GB HBM3 | HBM4 (capacity per partner configuration) |
| Memory Bandwidth | 3.35 TB/s | – |
| Compute | ||
| FP32 | 67 TFLOPS | – |
| FP64 Tensor | 67 TFLOPS | – |
| TF32 Tensor | 989 TFLOPS | – |
| FP16/BF16 Tensor | 1,979 TFLOPS | – |
| FP8 Tensor | 3,958 TFLOPS | – |
| INT8 Tensor | 3,958 TOPS | – |
| Platform | ||
| Interface | Gen5 128 GB/s | – |
| NVLink | 900 GB/s (4th Gen) | NVLink 6, PCIe Gen 6 |
| TDP | Up to 700W (configurable) | – |
| Form Factor | SXM | SXM / rack-integrated |
| Multi-Instance GPU (MIG) | Up to 7 instances (10GB each) | – |
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.