YOLO26n — ONNX

ONNX export of YOLO26n, the nano variant of Ultralytics YOLO26 object detection model. Features built-in NMS (no post-processing required). Trained on COCO with 80-class output.

Mirrored for use with inference4j, an inference-only AI library for Java.

Original Source

• Repository: onnx-community (Ultralytics YOLO26)
• License: agpl-3.0

Usage with inference4j

try (Yolo26 model = Yolo26.fromPretrained("models/yolo26n")) {
    List<Detection> detections = model.detect(Path.of("street.jpg"));
    detections.forEach(d -> System.out.printf("%s (%.0f%%) at [%.0f, %.0f, %.0f, %.0f]%n",
        d.label(), d.confidence() * 100,
        d.box().x1(), d.box().y1(), d.box().x2(), d.box().y2()));
}

Model Details

Property	Value
Architecture	YOLO26 Nano (NMS-free single-shot detector)
Task	Object detection (COCO 80 classes)
Input	[1, 3, 640, 640] — RGB, normalized 0-1
Output	NMS-free — outputs filtered detections directly
Post-processing	None required (built-in NMS)
Original framework	PyTorch (Ultralytics)

License

This model is licensed under the AGPL-3.0 License. Original model by Ultralytics, exported by onnx-community.