NVIDIA Vera Rubin Superchip vs NVIDIA Rubin CPX
Side-by-side specifications, performance, and rental pricing for datacenter AI workloads.
Vera Rubin: Rubin, 2026 Rubin CPX: Rubin, 2026
VRAM
576 vs 128 GB
448 GB more for the Vera Rubin
NVIDIA Vera Rubin Superchip
Full specs →Architecture
Rubin
Year
2026
VRAM
576 GB
NVIDIA Rubin CPX
Full specs →Architecture
Rubin
Year
2026
VRAM
128 GB
Specifications
Vera Rubin Rubin CPX
Architecture
RubinRubin
Launch Year
20262026
Form Factor
Superchip—
VRAM
576 GB128 GB
Memory Bandwidth
44.0 TB/s—
TDP
——
Process Node
——
Spec Vera Rubin Rubin CPX
Architecture RubinRubin
Launch Year 20262026
Form Factor Superchip—
VRAM 576 GB128 GB
Memory Bandwidth 44.0 TB/s—
TDP ——
Process Node ——
Performance (TFLOPS)
Vera Rubin Rubin CPX
FP64
67.0 TFlops —
FP32
260.0 TFlops —
FP16
8000.0 TFlops
sparse not published
—BF16
8000.0 TFlops
sparse not published
—FP8
35000.0 TFlops
sparse not published
—INT8
500.0 TFlops
sparse not published
—Tensor Core
FP64 (TC)
400.0 TFlops —
Precision Vera Rubin Rubin CPX
FP64 67.0 TFlops —
FP32 260.0 TFlops —
FP16 8000.0 TFlops
sparse not published
—BF16 8000.0 TFlops
sparse not published
—FP8 35000.0 TFlops
sparse not published
—INT8 500.0 TFlops
sparse not published
—Tensor Core
FP64 (TC) 400.0 TFlops —
FLOPS by Precision
What actually differs
Both GPUs launched in 2026: the Vera Rubin on Rubin and the Rubin CPX on Rubin.
Memory is 576 GB against 128 GB. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.
For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.
Frequently asked questions
Which has more memory, the Vera Rubin or the Rubin CPX?
The Vera Rubin carries 576 GB of VRAM versus 128 GB for the Rubin CPX.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.