Skip to content
imaged

Object detection

YOLO11n

The 2024 Ultralytics nano detector, the recommended live-feed default.

AGPL-3.0
Explore model

About this model

YOLO11n is the nano-size detector from the 2024 Ultralytics release. It returns COCO 80 detections with class, score, and box. The ONNX export is around 10 MB and includes NMS in the graph.

On M1 Max with CoreML, warm runs land near 46 ms, about 22 fps. The same family also loads a license-plate fine-tune of YOLO11n that drives the LPR pipeline. Both checkpoints share the same parser.

What you can build

  • Run live person and vehicle detection on a phone or laptop.
  • Feed detections into ByteTrack or DeepSort for multi-object tracking.
  • Detect license plates with the bundled fine-tune before OCR.
  • Pair with the SCRFD face detector in a multi-stage pipeline.

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

Input
A London street with a bus and pedestrians.
Output
The bus scene with detection boxes around the bus and four people.

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

71.1 ms

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 100.85

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

45.8 ms

Warm inference

Catalog reference · variant unspecifiedCold start (ms): 522.75

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/yolo11n.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;