YOLO v11

Ultralytics YOLO11 delivers state-of-the-art real-time vision performance with 22% fewer parameters and higher accuracy than previous generations.

YOLO11 streamlines computer vision by combining a redesigned backbone with an optimized neck for superior feature extraction. It supports five core tasks: detection, instance segmentation, oriented object detection (OBB), pose estimation, and classification. The model achieves a higher mAP on the COCO dataset while reducing computational load, making it ideal for edge devices (like the NVIDIA Jetson). You get faster inference and easier deployment through the unified Ultralytics Python framework.

https://docs.ultralytics.com/models/yolo11/