Install

pip install ultralytics

Development

git clone https://github.com/ultralytics/ultralytics
cd ultralytics
pip install -e .

Usage

1. CLI

To simply use the latest Ultralytics YOLO models

yolo task=detect    mode=train    model=yolov8n.yaml      args=...
          classify       predict        yolov8n-cls.yaml  args=...
          segment        val            yolov8n-seg.yaml  args=...
                         export         yolov8n.pt        format=onnx

2. Python SDK

To use pythonic interface of Ultralytics YOLO model

from ultralytics import YOLO

model = YOLO("yolov8n.yaml")  # create a new model from scratch
model = YOLO(
    "yolov8n.pt"
)  # load a pretrained model (recommended for best training results)
results = model.train(data="coco128.yaml", epochs=100, imgsz=640)
results = model.val()
results = model.predict(source="bus.jpg")
success = model.export(format="onnx")

Models

Model	size (pixels)	mAPval 50-95	Speed CPU (ms)	Speed T4 GPU (ms)	params (M)	FLOPs (B)
YOLOv5n	640	28.0	-	-	1.9	4.5
YOLOv6n	640	35.9	-	-	4.3	11.1
YOLOv8n	640	37.5	-	-	3.2	8.9
YOLOv5s	640	37.4	-	-	7.2	16.5
YOLOv6s	640	43.5	-	-	17.2	44.2
YOLOv8s	640	44.7	-	-	11.2	28.8
YOLOv5m	640	45.4	-	-	21.2	49.0
YOLOv6m	640	49.5	-	-	34.3	82.2
YOLOv8m	640	50.3	-	-	25.9	79.3
YOLOv5l	640	49.0	-	-	46.5	109.1
YOLOv6l	640	52.5	-	-	58.5	144.0
YOLOv7	640	51.2	-	-	36.9	104.7
YOLOv8l	640	52.8	-	-	43.7	165.7
YOLOv5x	640	50.7	-	-	86.7	205.7
YOLOv7-X	640	52.9	-	-	71.3	189.9
YOLOv8x	640	53.7	-	-	68.2	258.5
YOLOv5x6	1280	55.0	-	-	140.7	839.2
YOLOv7-E6E	1280	56.8	-	-	151.7	843.2
YOLOv8x6 +TTA	1280	- -	- -	- -	97.4	1047.2 -

If you're looking to modify YOLO for R&D or to build on top of it, refer to Using Trainer Guide on our docs.