The Procyon AI benchmark measures inference times on popular models like MobileNet, ResNet, and YOLO. It's designed to show how well a system can accelerate machine learning tasks in real use.
In UL Procyon AI Computer Vision, the A7 Max scores 100 overall, with 1.13 in MobileNet V3, 14.02 in ResNet 50, 37.95 in Inception V4, 50.92 in DeepLab V3, 91.98 in YOLO V3, and 5,394.91 in REAL-ESRGAN. That leaves it near the middle overall, with DeepLab V3 looking the best and MobileNet V3 being one of the weaker results.