| Customization: | Available |
|---|---|
| Application: | Enterprise Level |
| Instruction System: | Cuda Ptx |
Suppliers with verified business licenses
Audited by an independent third-party inspection agency


Architecture | Ampere Architecture |
CUDA Cores | 9216 CUDA Cores |
Memory | 24GB GDDR6 |
Memory Bandwidth | 600GB/s |
Tensor Cores | 3rd-gen Tensor Cores (FP16/TF32/BFLOAT16/INT8 precision) |
RT Cores | 72 RT Cores |
PCIe Interface | PCIe 4.0 x16 |
TDP | 150W |
FP Performance | FP32: 31.2 TFLOPS FP16: 125 TFLOPS |
Video Codec | AV1 Decoding Support 1 Encoder + 2 Decoders |
Virtualization | NVIDIA vGPU Technology |
Form Factor | Single-slot Full-Height Full-Length (FHFL) |
Warranty | 3-Year Official Warranty |
Applications | |
AI Inference/Training | High-performance AI inference and training for large models |
Graphics Rendering | Virtual Desktop Infrastructure (VDI), 3D modeling & rendering |
HPC | Accelerates data analysis and high-throughput computing |
Multi-Tenancy | Multi-user resource sharing via vGPU partitioning |





