GPU Architecture |
NVIDIA Ada Lovelace Architecture |
CUDA Cores |
18,176 |
Tensor Cores |
568 |
RT Cores |
142 |
Single-Precision Performance1 |
91.1 TFLOPS |
RT Core Performance1 |
210.6 TFLOPS |
Tensor Performance1 |
1457.0 TFLOPS2 |
GPU Memory |
48 GB GDDR6 with ECC |
Memory Interface |
384-bit |
Memory Bandwidth |
960 GB/s |
Max Power Consumption |
300W |
Thermal Solution |
Blower Active Fan |
Graphics Bus |
PCI 4.0 x16 |
Display Connectors |
DP 1.4 (4)3 |
Form Factor |
4.4” H x 10.5” L Dual Slot |
Product Weight |
1.180 kg |
1Peak rates are based on GPU boost clock.
2Effective FP8 TFLOPS using the new sparsity feature.
3Display ports are on by default for the RTX 6000 Ada Generation. Display ports are not active when using vGPU software. Virtualization support for the RTX 6000 Ada Generation will be available in an upcoming NVIDIA virtual GPU (vGPU) release, anticipated in Q1, 2023.