GPU Architecture | NVIDIA Volta |
NVIDIA Tensor Cores | 640 |
NVIDIA CUDA ® Cores | 5120 |
Double-Precision Performance | 7 TFLOPS |
Single-Precision Performance | 14 TFLOPS |
Tensor Performance | 112 TFLOPS |
Memory Size | 32GB/16GB HBM2 |
Memory Bandwidth | 900 GB/sec |
Memory Interface | 4096-bit |
System Interface | PCIe Gen3 |
Form Factor | PCIe Full Height/Length |
Thermal Solution | Passive |
Max Power Consumption | 250 W |
Compute APIs | CUDA, DirectCompute, OpenCL ™ , OpenACC |