YOLOv8

yolov8

External baseline · 2023

Variants5

Parameters3.2M - 68.2M

Best Paper mAP53.9%

Licensenon-permissive

Model Variants

All variants benchmarked on COCO val2017. Click any variant for the full hardware breakdown.

Model	Params	GFLOPs	Paper mAP	Best FPS
YOLOv8-N	3.2M	8.7	37.3%	81 FPS Raspberry Pi 5 · HailoRT INT8	Details
YOLOv8-S	11.2M	28.6	44.9%	68 FPS Raspberry Pi 5 · HailoRT INT8	Details
YOLOv8-M	25.9M	78.9	50.2%	27 FPS Raspberry Pi 5 · HailoRT INT8	Details
YOLOv8-L	43.7M	165.2	52.9%	19 FPS NVIDIA Jetson Orin Nano Super 8GB · TensorRT FP16	Details
YOLOv8-X	68.2M	257.8	53.9%	14 FPS NVIDIA Jetson Orin Nano Super 8GB · TensorRT FP16	Details

Architecture

Type

one-stage

Backbone

CSPDarknet

Neck

C2f-PAFPN

Head

Decoupled

anchor-freenmsTrained on COCO train2017

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