Before You Start

Start from a Python>=3.8 environment with PyTorch>=1.7 installed. To install PyTorch see To install YOLOv5 dependencies:

pip install -qr  # install dependencies

Model Description

YOLOv5 Models  

YOLOv5 is a family of compound-scaled object detection models trained on the COCO dataset, and includes simple functionality for Test Time Augmentation (TTA), model ensembling, hyperparameter evolution, and export to ONNX, CoreML and TFLite.

Model size APval APtest AP50 SpeedV100 FPSV100   params GFLOPS
YOLOv5s 640 36.8 36.8 55.6 2.2ms 455   7.3M 17.0
YOLOv5m 640 44.5 44.5 63.1 2.9ms 345   21.4M 51.3
YOLOv5l 640 48.1 48.1 66.4 3.8ms 264   47.0M 115.4
YOLOv5x 640 50.1 50.1 68.7 6.0ms 167   87.7M 218.8
YOLOv5x + TTA 832 51.9 51.9 69.6 24.9ms 40   87.7M 1005.3

YOLOv5 Performance ** GPU Speed measures end-to-end time per image averaged over 5000 COCO val2017 images using a V100 GPU with batch size 32, and includes image preprocessing, PyTorch FP16 inference, postprocessing and NMS. EfficientDet data from google/automl at batch size 8.

Load From PyTorch Hub

This simple example loads a pretrained YOLOv5s model from PyTorch Hub as model and passes two image URLs for batched inference.

import torch

# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True)

# Images
dir = ''
imgs = [dir + f for f in ('zidane.jpg', 'bus.jpg')]  # batched list of images

# Inference
results = model(imgs)

# Results
results.print()  # or .show()

# Data
print(results.xyxy[0])  # print img1 predictions (pixels)
#                   x1           y1           x2           y2   confidence        class
# tensor([[7.50637e+02, 4.37279e+01, 1.15887e+03, 7.08682e+02, 8.18137e-01, 0.00000e+00],
#         [9.33597e+01, 2.07387e+02, 1.04737e+03, 7.10224e+02, 5.78011e-01, 0.00000e+00],
#         [4.24503e+02, 4.29092e+02, 5.16300e+02, 7.16425e+02, 5.68713e-01, 2.70000e+01]])

For YOLOv5 PyTorch Hub inference with PIL, OpenCV, Numpy or PyTorch inputs please see the full YOLOv5 PyTorch Hub Tutorial.




Issues should be raised directly in the repository. For business inquiries or professional support requests please visit or email Glenn Jocher at