--- library_name: pytorch license: other tags: - real_time - android pipeline_tag: object-detection --- ![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/yolov11_det/web-assets/model_demo.png) # YOLOv11-Detection: Optimized for Qualcomm Devices Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes and classes of objects in an image. This is based on the implementation of YOLOv11-Detection found [here](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect). This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_det) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary). Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) to run these models on a hosted Qualcomm® device. ## Getting Started Due to licensing restrictions, we cannot distribute pre-exported model assets for this model. Use the [Qualcomm® AI Hub Models](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_det) Python library to compile and export the model with your own: - Custom weights (e.g., fine-tuned checkpoints) - Custom input shapes - Target device and runtime configurations See our repository for [YOLOv11-Detection on GitHub](https://github.com/quic/ai-hub-models/blob/main/qai_hub_models/models/yolov11_det) for usage instructions. ## Model Details **Model Type:** Model_use_case.object_detection **Model Stats:** - Model checkpoint: YOLO11-N - Input resolution: 640x640 - Number of parameters: 2.64M - Model size (float): 10.1 MB - Model size (w8a8): 2.83 MB - Model size (w8a16): 3.30 MB ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | YOLOv11-Detection | ONNX | float | Snapdragon® X Elite | 6.595 ms | 5 - 5 MB | NPU | YOLOv11-Detection | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.471 ms | 1 - 261 MB | NPU | YOLOv11-Detection | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.088 ms | 5 - 9 MB | NPU | YOLOv11-Detection | ONNX | float | Qualcomm® QCS9075 | 6.718 ms | 5 - 7 MB | NPU | YOLOv11-Detection | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.036 ms | 0 - 227 MB | NPU | YOLOv11-Detection | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.672 ms | 1 - 220 MB | NPU | YOLOv11-Detection | ONNX | float | Snapdragon® X2 Elite | 3.021 ms | 3 - 3 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® X Elite | 5.769 ms | 2 - 2 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 3.275 ms | 0 - 219 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Qualcomm® QCS6490 | 324.68 ms | 95 - 100 MB | CPU | YOLOv11-Detection | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 5.275 ms | 2 - 8 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Qualcomm® QCS9075 | 5.965 ms | 2 - 5 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Qualcomm® QCM6690 | 166.046 ms | 95 - 104 MB | CPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.364 ms | 0 - 72 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 146.827 ms | 95 - 105 MB | CPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.162 ms | 0 - 81 MB | NPU | YOLOv11-Detection | ONNX | w8a16 | Snapdragon® X2 Elite | 2.33 ms | 0 - 0 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® X Elite | 13.783 ms | 8 - 8 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 8.756 ms | 9 - 406 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® QCS6490 | 134.357 ms | 62 - 71 MB | CPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 12.658 ms | 8 - 13 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® QCS9075 | 18.42 ms | 8 - 11 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Qualcomm® QCM6690 | 79.574 ms | 54 - 65 MB | CPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 6.863 ms | 6 - 333 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 71.799 ms | 65 - 77 MB | CPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 6.382 ms | 9 - 348 MB | NPU | YOLOv11-Detection | ONNX | w8a8_mixed_int16 | Snapdragon® X2 Elite | 6.82 ms | 10 - 10 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Snapdragon® X Elite | 4.263 ms | 5 - 5 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 2.894 ms | 5 - 215 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 3.905 ms | 2 - 13 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Qualcomm® SA8775P | 5.28 ms | 0 - 183 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Qualcomm® QCS9075 | 5.043 ms | 5 - 11 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 8.408 ms | 5 - 196 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Qualcomm® SA8295P | 8.623 ms | 0 - 166 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.251 ms | 5 - 190 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.864 ms | 5 - 188 MB | NPU | YOLOv11-Detection | QNN_DLC | float | Snapdragon® X2 Elite | 2.564 ms | 5 - 5 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® X Elite | 4.644 ms | 2 - 2 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 2.977 ms | 0 - 200 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 8.026 ms | 2 - 176 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 4.279 ms | 2 - 4 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® SA8775P | 4.856 ms | 0 - 177 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 4.804 ms | 0 - 4 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 18.509 ms | 2 - 181 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Qualcomm® SA7255P | 8.026 ms | 2 - 176 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 2.076 ms | 2 - 178 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 4.557 ms | 0 - 179 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 1.771 ms | 2 - 184 MB | NPU | YOLOv11-Detection | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 2.232 ms | 2 - 2 MB | NPU | YOLOv11-Detection | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 2.617 ms | 0 - 110 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 12.263 ms | 0 - 82 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 3.624 ms | 0 - 6 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® SA8775P | 5.011 ms | 0 - 92 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® QCS9075 | 4.717 ms | 0 - 13 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 7.151 ms | 0 - 78 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® SA7255P | 12.263 ms | 0 - 82 MB | NPU | YOLOv11-Detection | TFLITE | float | Qualcomm® SA8295P | 8.014 ms | 0 - 52 MB | NPU | YOLOv11-Detection | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 1.958 ms | 0 - 89 MB | NPU | YOLOv11-Detection | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 1.534 ms | 0 - 89 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.111 ms | 0 - 55 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCS6490 | 3.754 ms | 0 - 6 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 3.439 ms | 0 - 39 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.633 ms | 0 - 2 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® SA8775P | 2.083 ms | 0 - 43 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCS9075 | 2.008 ms | 0 - 6 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCM6690 | 7.626 ms | 0 - 165 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 1.874 ms | 0 - 55 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® SA7255P | 3.439 ms | 0 - 39 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Qualcomm® SA8295P | 2.472 ms | 0 - 38 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.826 ms | 0 - 42 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.712 ms | 0 - 155 MB | NPU | YOLOv11-Detection | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.731 ms | 0 - 43 MB | NPU ## License * The license for the original implementation of YOLOv11-Detection can be found [here](https://github.com/ultralytics/ultralytics/blob/main/LICENSE). ## References * [Ultralytics YOLOv11 Docs: Object Detection](https://docs.ultralytics.com/tasks/detect/) * [Source Model Implementation](https://github.com/ultralytics/ultralytics/tree/main/ultralytics/models/yolo/detect) ## Community * Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI. * For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).