Here are the minimum system requirements for the UL Procyon AI Image Generation Benchmark tests.
Please note that the different inference engines used in the tests have different System requirements.
| OS | Windows 10, 64-bit or Windows 11 |
| Processor | 2 GHz dual-core CPU |
| Memory | 16 GB |
| Storage | 20 GB (75 GB recommended) |
Stable Diffusion System Requirements
Stable Diffusion XL
| TensorRT | NVIDIA RTX GPU with 10GB VRAM |
| OpenVINO | Intel Arc discrete GPU with 16GB VRAM |
| ONNX Runtime | 16GB VRAM |
Stable Diffusion 1.5 (FP16)
| TensorRT | NVIDIA RTX GPU with 10GB VRAM |
| OpenVINO | Intel Arc discrete GPU with 8GB VRAM or Intel integrated GPU with 32GB system RAM |
| ONNX Runtime | 8GB VRAM (Discrete GPU) 32GB RAM (Integrated GPU) |
Stable Diffusion 1.5 (INT8)
| TensorRT | NVIDIA 30 series GPU or later |
| OpenVINO | An Intel NPU, Intel integrated GPU or Intel Arc discrete GPU |
Ready for Windows 11
UL Procyon benchmarks are compatible with Windows 11.