Ensuring safety compliance in manufacturing is recognized to pose considerable challenges. Owing to the unreliability of human observations in comparison with machine-based systems, intelligent systems are increasingly employed to support human efforts in various industrial applications. An additional set of trained models for YOLOv11 is presented in this study for a custom open-sourced dataset published by our research group, which is utilized to identify standard safety compliance object classes. The performance of the newest model in the YOLOv11 series is evaluated and compared with that of its predecessors, YOLOv9 and YOLOv10. The objective is to render these models readily available for other researchers and consumers, thereby advancing accessibility and equity for safety compliance using computer vision.