Architecture |
Ada Lovelace |
Memory Interface |
256 bit |
GPU memory |
32GB GDDR6 |
Memory Bandwidth |
576GB/s |
Error Correcting Code (ECC) |
Yes |
NVIDIA Ada Lovelace Architecture-based CUDA Cores |
12,800 |
NVIDIA 4th Gen Tensor Cores |
400 |
NVIDIA 3rd Gen RT Cores |
100 |
Single-Precision Performance |
65.3 TFLOPS Peak |
RT Core Performance |
151.0 TFLOPS Peak |
Tensor Performance |
1044.4 TFLOPS with sparsity |
System Interface |
PCI3 4.0 x 16 |
Power Consumption |
Total board power: 250W |
Thermal Solution |
Active |
Form Factor |
4.4" (H) x 9.5" (L) |
|
Dual-slot |
Display Connectors |
4 * DisplayPort 1.4a (on by default) |
|
Display ports aren't active when using vGPU software. |
Max Simultaneous Displays |
4 * 4096 x 2160 @120Hz |
|
4 * 5120 x 2880 @ 60Hz |
|
2 * 7680 x 4320 @ 60Hz |
Encode/Decode |
2 * Encode |
|
2 * Decode |
|
+AV1 encode and decode |
VR Ready |
Yes |
vGPU Software Support |
NVIDIA vPC/vApps |
|
NVIDIA RTX Virtual Workstation |
vGPU Profiles Supported |
For the full list of vGPU supported profiles, visit: https://www.nvidia.com/content/dam/en-zz/Solutions/design-visualization/solutions/resources/documents1/Virtual-GPU-Packaging-and-Licensing-Guide.pdf |
Graphics APIs |
Direct X 12 |
|
Shader Model 6.7 |
|
OpenGL 4.6 (Based on Khronos specification) |
|
Vulkan 1.3 (Based on Khronos specification) |
Compute APIs |
CUDA 12.2 |
|
OpenCL 3.0 |
|
DirectCompute |
NVIDIA NVLink® |
No |