YOLO11

DeGirum's YOLO11 models consists of a variety of models gathered using ultralytics repository. The models are located at OPENVINO models' zoo.

COCO

Dataset Name	Training Images	Validation Images	Class Labels	License
COCO	117,266	5,000	80 classes	Annotations under CC BY 4.0

Model Architecture	Input Size	Precision	Runtime	Device Type	mAP 50-95	mAP 50	mAP Type	FPS
yolo11n_silu_coco	640x640	INT8	OPENVINO	NPU	0.3913	0.5492	bbox	90.8
yolo11n_silu_coco	640x640	FP32	OPENVINO	NPU	0.3931	0.5513	bbox	105.0
yolo11s_silu_coco	640x640	INT8	OPENVINO	NPU	0.4649	0.6337	bbox	49.6
yolo11s_silu_coco	640x640	FP32	OPENVINO	NPU	0.4663	0.6347	bbox	52.1
yolo11m_silu_coco	640x640	INT8	OPENVINO	NPU	0.5116	0.6794	bbox	24.7
yolo11m_silu_coco	640x640	FP32	OPENVINO	NPU	0.5132	0.6818	bbox	21.7
yolo11l_silu_coco	640x640	INT8	OPENVINO	NPU	0.5293	0.6965	bbox	20.5
yolo11l_silu_coco	640x640	FP32	OPENVINO	NPU	0.5318	0.6988	bbox	16.8
yolo11x_silu_coco	640x640	INT8	OPENVINO	NPU	0.5451	0.7145	bbox	11.7
yolo11x_silu_coco	640x640	FP32	OPENVINO	NPU	0.545	0.7135	bbox	7.6

COCO-Pose

Dataset Name	Training Images	Validation Images	Class Labels	License
COCO Pose Keypoints	56,599	5,000	0: Person 17 keypoints	Annotations under CC BY 4.0

Here is an example of the output for 17 pose keypoints.

Model Architecture	Input Size	Precision	Runtime	Device Type	mAP 50-95	mAP 50	mAP Type	FPS
yolo11n_silu_coco_pose	640x640	INT8	OPENVINO	NPU	0.5232 0.4897	0.7057 0.8052	bbox keypoints	82.2
yolo11n_silu_coco_pose	640x640	FP32	OPENVINO	NPU	0.5224 0.4986	0.7051 0.8075	bbox keypoints	89.5
yolo11s_silu_coco_pose	640x640	INT8	OPENVINO	NPU	0.5549 0.574	0.7228 0.8607	bbox keypoints	46.3
yolo11s_silu_coco_pose	640x640	FP32	OPENVINO	NPU	0.556 0.585	0.7263 0.86	bbox keypoints	47.0
yolo11m_silu_coco_pose	640x640	INT8	OPENVINO	NPU	0.5749 0.6378	0.7321 0.8901	bbox keypoints	23.4
yolo11m_silu_coco_pose	640x640	FP32	OPENVINO	NPU	0.5776 0.6491	0.7337 0.8911	bbox keypoints	20.5
yolo11l_silu_coco_pose	640x640	INT8	OPENVINO	NPU	0.577 0.652	0.7313 0.8949	bbox keypoints	19.4
yolo11l_silu_coco_pose	640x640	FP32	OPENVINO	NPU	0.5794 0.6621	0.7332 0.8985	bbox keypoints	16.0
yolo11x_silu_coco_pose	640x640	INT8	OPENVINO	NPU	0.5918 0.6855	0.744 0.9071	bbox keypoints	11.2
yolo11x_silu_coco_pose	640x640	FP32	OPENVINO	NPU	0.5922 0.6927	0.7447 0.9098	bbox keypoints	7.4

COCO-Segmentation

Dataset Name	Training Images	Validation Images	Class Labels	License
COCO	117,266	5,000	80 classes	Annotations under CC BY 4.0

Here is an example of the output for instance segmentation.

Model Architecture	Input Size	Precision	Runtime	Device Type	mAP 50-95	mAP 50	mAP Type	FPS
yolo11n_silu_coco_seg	640x640	INT8	OPENVINO	NPU	0.3856 0.3179	0.5396 0.5083	bbox mask	71.0
yolo11n_silu_coco_seg	640x640	FP32	OPENVINO	NPU	0.389 0.3217	0.5438 0.5128	bbox mask	75.5
yolo11s_silu_coco_seg	640x640	INT8	OPENVINO	NPU	0.4608 0.3762	0.6273 0.5933	bbox mask	42.0
yolo11s_silu_coco_seg	640x640	FP32	OPENVINO	NPU	0.4637 0.3778	0.6304 0.5965	bbox mask	38.6
yolo11m_silu_coco_seg	640x640	INT8	OPENVINO	NPU	0.5115 0.4164	0.6794 0.6483	bbox mask	19.9
yolo11m_silu_coco_seg	640x640	FP32	OPENVINO	NPU	0.5132 0.4178	0.6818 0.6503	bbox mask	14.7
yolo11l_silu_coco_seg	640x640	INT8	OPENVINO	NPU	0.5293 0.4289	0.6965 0.6664	bbox mask	17.2
yolo11l_silu_coco_seg	640x640	FP32	OPENVINO	NPU	0.5318 0.4309	0.6988 0.6685	bbox mask	12.2
yolo11x_silu_coco_seg	640x640	INT8	OPENVINO	NPU	0.545 0.4413	0.7142 0.6824	bbox mask	8.4
yolo11x_silu_coco_seg	640x640	FP32	OPENVINO	NPU	0.5449 0.4405	0.7135 0.6822	bbox mask	4.9