This folder contains a minimal example of running YOLO26 (End-to-End) using ONNX Runtime.

Unlike traditional YOLO models that require complex post-processing (Non-Maximum Suppression), YOLO26 outputs a [1, 300, 6] tensor directly. This means the model has already selected the best detections for us.

(The model directly outputs these boxes without NMS)

• NMS_Free_Inference_with_YOLO26.ipynb: The main notebook. It loads the ONNX model, pre-processes the image, and visualizes the specific output format.
• Person_Dog.png: The raw input image used for testing (with class labels).
• YOLO26_Test_image.png: The raw input image used for testing.
• YOLO26_Test_image.png: Side-by-side comparison showing the original input (left) and the YOLO26 detections (right).

LearnOpenCV is maintained by BigVision.AI, a computer vision and AI consulting company. We help organizations design, build, optimize, and deploy production-ready AI solutions. Our team has deep expertise in computer vision, deep learning, multimodal AI, and edge deployment, with experience solving complex technical challenges across industries.

Have a project in mind? Talk with our expert AI solution builders.