ACEMAGIC Tank M1A Pro+ Review 23

ACEMAGIC Tank M1A Pro+ Review

Storage Performance »

AI Performance

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-world use.


The Tank M1A Pro+ tops the UL Procyon AI Computer Vision chart with an overall score of 220, ahead of the NZXT H2 at 202 and well in front of the next group, which starts with the GEEKOM A9 Max at 137. It also leads every individual model test here, posting 0.79 ms in MobileNet V3, 6.18 ms in ResNet 50, 17.05 ms in Inception V4, 21.56 ms in DeepLab V3, 37.87 ms in YOLO V3, and 1994.93 ms in Real-ESRGAN, with the NZXT H2 trailing in each case at 0.93 ms, 6.59 ms, 17.33 ms, 23.27 ms, 40.32 ms, and 2245.44 ms respectively. That is an outstanding result for a platform this compact, especially since it finishes ahead not only of the rest of the Mini-PC field, but also of a system built around an RTX 5080.
Next Page »Storage Performance
View as single page
Nov 20th, 2026 15:29 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts