Flopper.io GPU Specification Sheet
October 16, 2026
flopper.io/gpu/nvidia-vera-rubin-superchipNVIDIA Vera Rubin Superchip
Rubin Superchip 2026
Overview
260.0
FP32 TFLOPS
576 GB
Memory HBM4
44.0 TB/s
Memory Bandwidth
—
TDP
Performance Metrics
| Precision | Peak (dense) | 2:4 Sparse | Bits | Type |
|---|---|---|---|---|
| FP64 | 67.0 | N/A | 64 | Standard |
| FP64 Tensor | 400.0 | N/A | 64 | Tensor Core |
| FP32 | 260.0 | N/A | 32 | Standard |
| FP32 Tensor | 800.0 | N/A | — | Tensor Core |
| TF32 | 4000.0 | not published | 32 | Standard |
| BF16 | 8000.0 | not published | 16 | Standard |
| FP16 | 8000.0 | not published | 16 | Standard |
| FP8 | 35000.0 | not published | 8 | Standard |
| FP6 | 35000.0 | not published | — | Standard |
| FP4 | — | not published | — | Standard |
| NVFP4 | 100000.0 | not published | — | Standard |
| INT8 | 500.0 | not published | 8 | Standard |
Memory Specifications
- Capacity
- 576 GB
- Type
- HBM4
- Bandwidth
- 44.0 TB/s
- Interface Width
- —
Power & Efficiency
- TDP
- —
- Max power
- —
- FP32 dense TFLOPS per W of TDP
- —
- FP16 dense TFLOPS per W of TDP
- —
Per W of TDP figures divide dense peak by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead.
Hardware Details
Vendor: NVIDIA
Architecture: Rubin
Process Node: —
Form Factor: Superchip
Launch Year: 2026
Variant: Superchip
Interconnect
GPU-to-GPU: NVLink-C2C
Bandwidth: 1.8 TB/s bidirectional, per GPU
Enterprise Features
Sparsity: Yes