Architecture | Ampere |
Memory Interface | 5,120 bit |
GPU memory | 40GB HBM2 |
Memory Bandwidth | 1.5TB/s |
CUDA® Cores | 6,912 |
Tensor Cores | 432 |
Double-Precision Performance | 9.7 TFLOPS |
Single-Precision Performance | 19.5 TFLOPS |
Peak Tensor Performance | 632.8 TFLOPS |
Multi-Instance GPU | Up to 7 MIG instances @ 5GB |
NVIDIA® NVLINK® | Yes |
NVLink® Bandwidth | 400GB/s |
Graphics Bus | PCIe 4.0 x 16 |
Power Consumption | 240W |
Thermal | Active |
Form Factor | 4.4" (H) x 10.5" (L) |
Dual-slot |