Vision Analysis

LeaderboardNVIDIA Jetson Orin Nano Super 8GB · TensorRT FP16As of Aug 9, 2026

#	Model	Family	mAP	FPS	Latency	Params
1	D-FINE-X	dfine	61.4%	14.2	70.5ms	62.0M
2	ECDet-X	ec	61.0%	5.7	174.8ms	49.0M
3	DEIMv2-X	deimv2	60.8%	5.4	184.5ms	50.3M
4	ECDet-L	ec	60.0%	6.2	161.5ms	31.0M
5	D-FINE-L	dfine	60.0%	18.6	53.6ms	31.0M
6	RT-DETRv4-X	rtdetrv4	60.0%	13.8	72.7ms	62.0M
7	DEIM-X	deim	59.6%	14.1	70.9ms	62.0M
8	RF-DETR-L	rfdetr	58.7%	8.4	119.3ms	128.0M
9	DEIMv2-L	deimv2	58.5%	6.2	160.8ms	32.2M
10	ECDet-M	ec	58.3%	8.3	119.7ms	18.0M

COCO val2017 · ranked by mAP@50-95visionanalysis.org

Run any model with one line

LibreYOLO has the best catalogue of state-of-the-art detectors, all behind one MIT-licensed Python API.

from libreyolo import LibreYOLO, SAMPLE_IMAGE # LibreYOLO has the best catalogue of state-of-the-art models. model = LibreYOLO("LibreRFDETRl.pt") # RF-DETR-L (transformer flagship) results = model(SAMPLE_IMAGE, save=True) # run inference, save the annotated image # Swap in any other model, same one-line API (weights auto-download): # LibreYOLO("LibreYOLO9c.pt") # YOLO9-C # LibreYOLO("LibreYOLOXx.pt") # YOLOX-X # LibreYOLO("LibreDFINEx.pt") # D-FINE-X # LibreYOLO("LibreRTDETRr50.pt") # RT-DETR-R50