Object detection

YOLOv8n

The 2023 Ultralytics nano detector, kept as a stable baseline.

AGPL-3.0

Explore model

About this model

YOLOv8n is the nano-size detector from the 2023 Ultralytics release. It outputs class, confidence, and bounding box for each object on COCO 80 classes. The build is small, around 12 MB on disk, and runs on every supported platform.

The SDK ships an ONNX export with NMS baked into the graph. Many existing pipelines were tuned against v8, so this checkpoint is included for compatibility and regression testing alongside YOLOv11 and YOLO26x.

What you can build

Detect people and vehicles in CCTV frames at 14 fps on M1 CPU.
Run a stable detector on existing pipelines tuned against v8 weights.
Pre-screen frames before a heavier YOLO26x pass for accuracy.
Drive a tracking pipeline that needs a small, predictable detector.

Variants & downloads

Choose the build for your app

Model files are being prepared.

Explore the model’s examples, source links, and integration notes while download variants are added.

Example inference

A London street with a bus and pedestrians, the canonical YOLO test image. — Input

The same scene with bounding boxes around the bus, four people, and a stop sign. — Output

Performance

Results depend on the device, runtime, and test conditions. Each measurement keeps its own context.

M1 Max, macOS 15, ONNX Runtime 1.17

See original catalog hardware notes · CPU

72.7 ms

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 74.91

Reference result from the existing catalog; conversion variant and measurement date were not recorded.

M1 Max, macOS 15, ONNX Runtime 1.17

See original catalog hardware notes · CoreML

42.6 ms

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 327.09

Reference result from the existing catalog; conversion variant and measurement date were not recorded.

Integration examples

Vision SDK integration

cpp

#include "imaged/sdk.hpp"
#include "imaged/families/yolo/yolo_family.hpp"

imaged::SDK::AI ai;
auto& config = ai.getOptions();
config.set("yolo.model", "weights/yolov8n.onnx");
config.set("yolo.confidenceThreshold", 0.25);

imaged::SDK::YoloFamily family;
family.load(config);

imaged::SDK::Image image;
image.load("frame.jpg");

auto result = family.run({&image});
auto& detections = std::get<imaged::SDK::DetectionResult>(result).detections;