This repository focuses on improving and extending the YOLO (You Only Look Once) family of real-time object detection models, including YOLO11, YOLOv8, YOLOv10, RT-DETR, YOLOv7, and YOLOv5. It provides modular customization options allowing users to swap or enhance key components such as the backbone, neck, head, loss functions, IoU metrics, and Non-Maximum Suppression (NMS) post-processing. Built on PyTorch and the Ultralytics framework, the project aims to make it easy for researchers and practitioners to experiment with and deploy improved detection architectures.